Skip to main content

Full text of "CMH Pub 9-2 Stilwell's Command Problems"

See other formats


China-Burma-India Theater 


WASHINGTON t £>.C, 1987 

First Printed 1956— CMH Pub 9-2 

For sale by the Superintendent of Documents, U.S. Government Printing Office 
Washington, D.C, 204G2 

Kent Roberts Greenfield, General Editor 

James P Baxter 
President, Williams College 

Gordon A, Craig 
Princeton University 

Elmer Ellis 
University of Missouri 

William T. Hutchinson 
University of Chicago 

Charles H. Taylor 
Harvard University 

Advisory Committee 

i\t of I June 

Maj. Gen. Elwyn D. Post 
Army Field Forces 

Bng. Gen. Verdi B Barnes 
Army War College 

Brig. Gen. C. E. Beauchamp 
Command and General Staff College 

Brig, Gen. Leonard J. Greeley 
Industrial College of the Armed Forces 

Col. Thomas D. Stamps 
United States Military Academy 

j. Gen Albert C 

Chief Histonan 
Chief, War Histories Division 
Chief, Editorial and Publication Division 
Chief, Editorial Branch 
Chief, Cartographic Branch 
Chief, Photographic Branch 

Kent Roberts Greenfield 
Col. G. G. O'Connor 
Lt. Col. T. E. Bennett 
Joseph R. Friedman 
Wsevolod AglaimorT 
Maj. Arthur T Lawry 

* Maj. Gen. Orlando Ward was succeeded by General Smith on i February 1953 



Although this is the second of a series covering the China-Burma-India 
Theater of Operations during World War II, it is a story in itself and one full 
of drama. The previous volume, Stilwell's Mission to China, recounts the early 
efforts of the United States to improve the combat efficiency of the Chinese 
Army. This second volume presents the problems of a commander, his staff, 
and his troops in a position so irregular and complex as to be unprecedented in 
U.S. Army history, and outlines the background of their position in Allied 
policy, military and political. Their position was determined by an arrangement 
among allies, one accepted with reservations by the War Department. There is 
a saying: "There is but one thing more difficult than fighting a war with 
Allies — this is to fight a war without them." 

A history of the Allied effort in China and Burma, to be complete, must be 
written in three dimensions, American-British-Chinese. The present volume 
is based on a thorough study of the American records and a wealth of informa- 
tion in General Stilwell's papers not previously explored. The full story of the 
war on the Asiatic mainland cannot be written until British studies are further 
advanced and the records and the views of the Chinese, of which only a super- 
ficial knowledge is now obtainable, have been disclosed and compared with 
those of their Western allies. Nevertheless, it seems desirable to tell the Amer- 
ican story now. It is needed to round out the history of our Army's global effort 
and to do justice to the Americans, high and low, who made their contribution 
to victory in a vast and difficult but at the time little-known theater. 

Washington, D. C. 
30 January 1953 

Maj. Gen., U.S.A. 
Chief of Military History 


The Authors 

From 1946 to 1953 the authors formed the CBI Section of the staff engaged 
in writing the history of the U.S. Army in World War II. 

Charles F. Romanus received the degree of Master of Arts in History at the 
University of Illinois in 1937, and pursued his work for the doctorate at 
Louisiana State University, where he was a teaching fellow in history. Entering 
the Army in 1943 he was commissioned in March 1945 and became a historical 
officer in the headquarters of the China Theater. He is now a historian in the 
Historical Section of the Office of the Quartermaster General, and is a captain 
in the USAR. 

Riley Sunderland graduated from the University of Chicago in 1937. In 
April 1942 he was called to active duty as a second lieutenant in the Field Artil- 
lery, and from July 1945 to May 1946 was in the Historical Section, Headquar- 
ters, India-Burma Theater, in New Delhi. He is now a senior operations 
research analyst for Technical Operations, Incorporated, of Arlington, 



When, in October 1943, Lt. Gen. Joseph W. Stilwell decided there was little 
more he personally could do to improve the combat efficiency of the Chinese 
Army, and decided to concentrate his efforts on the India-Burma scene, in 
effect his decision marked a change from the role of a staff officer, advising 
without the power to command, to that of a commander, giving orders. As 
Commanding General, United States Army Forces, China, Burma and India 
Theater of Operations, as acting Deputy Supreme Allied Commander, South- 
east Asia, as Commanding General, Chinese Army in India, and as Command- 
ing General, Northern Combat Area Command, Stilwell was charged with 
many duties. He was responsible for the active conduct of a campaign in north 
Burma, and for its immediate logistical support; for the execution of a number 
of projects agreed to by the President, the Prime Minister of Great Britain, and 
the Generalissimo of China to aid China; for the execution of projects ordered 
by the U.S. Joint Chiefs of Staff to reinforce U.S. operations in the Pacific, and 
for their logistical support; and, as deputy commander, for playing an appro- 
priate role in Southeast Asia Command should necessity demand. This multi- 
tude of roles, these often conflicting missions and requirements, meant that 
Stilwell faced a variety of command problems. This volume relates his efforts 
to solve them. 

The narrative, like that in the first volume of the subseries, Stilwell's Mission 
to China, is written at the level of the theater commander's headquarters or 
command post. Under this inherent limitation, it offers a contribution to an 
understanding of the American effort in China, Burma, and India in 1943 and 
1944, and to the study of Sino- American relations in the same period; it is not 
a definitive history of the war in Asia. The U.S. Air Force is telling its story in 
many volumes, written from its own point of view. So are the technical serv- 
ices. A British official history is well advanced. The Chinese may someday tell 
their story, and the beginnings of a Japanese official history are at hand. From 
these several sources, some historian may ultimately produce a fairly complete 
history of the war in Asia, which truly deserves to be called "the unknown 
war." This volume only continues what its predecessor began, that is, a recon- 
naissance of part of the area the future historian must cross. 

That the authors prepared the volume in this manner reflects primarily 


their mission— writing a history of the U.S. Army theater of operations in 
China, Burma, and India. It also reflects the nature of the documentary sources 
which the U.S. Army was able to open for them. 

The events of 1943-44 in China, Burma, and India, great though they were, 
were but the prelude to others. Looking back on 1943 and 1944, those who had 
seen the later events saw the prelude through the glass of their own experi- 
ences, and interpreted the actions and motives of the men of 1943 and 1944 
accordingly. But the men whose words and acts are recorded in these pages 
were no better at foretelling the future than men have ever been. Sometimes 
aware of what the future might hold, they were nonetheless preoccupied with 
the problems of the day, rather than those of next year. Standing on their side 
of the veil of the future, they saw through it dimly as men always see through 
it, not as men see the recent past. It is the historian's task to recreate the scene 
they saw, and let the reader judge them, as inevitably he will judge them, by 
the understanding of his own day. 

The volume has many references to disagreements between allies. Inevit- 
ably, the account of the disagreements takes up more space than the final 
agreement, though it is the latter which is the more important. The differences 
and disagreements are of interest to the student because they did happen, and 
because they are an inevitable part of coalition war. But they should not be 
magnified out of proportion. 

The authors are greatly indebted to many participants in the events 
described who were willing to offer comment and criticism on several drafts of 
the manuscript: Lt. Col. Charles E. Beach, Maj. Gen. Haydon L. Boatner, Brig. 
Gen. Frederic W. Boye, Brig. Gen. Waldemar F. Breidster, Col. Rothwell H. 
Brown, Col. Harry A. Buckley, Brig. Gen. Robert M. Cannon, Maj. Gen. Gil- 
bert X. Cheves, Lt. Col. George L. Converse, Maj. Gen. William E. R. Co veil, 
Maj. Gen. Howard C. Davidson, Col. Joseph K. Dickey, Brig. Gen. Frank 
Dorn, Lt. Col. Trevor N. Dupuy, Col. Ernest F. Easterbrook, Maj. Gen. Vernon 
Evans, Brig. Gen. Benjamin G. Ferris, Lt. Col. John B. George, Capt. Shy S. 
Greenspan, Maj. Gen. Thomas G. Hearn, Col. Francis Hill, Col. Charles N. 
Hunter, Maj. Gen. Patrick J. Hurley, Lt. Col. George T. Laughlin, Brig. Gen. 
Malcolm F. Lindsey, Col. Robert C. Lutz, Col. John E. McCammon, Lt. Col. 
George A. McGee, Jr., Col. Edward J. McNally, Lt. Col. Harry L. Mayfield, 
Maj. Gen. Frank D. Merrill, Lt. Col. William L. Osborne, Col. William R. 
Peers, Lt. Gen. Lewis A. Pick, Brig. Gen. Frank N. Roberts, Lt. Col. Joseph 
Rockis, General Brehon B. Somervell, Col. Carlos G. Spaht, Col. John H. 
Stodter, Lt. Gen. George E. Stratemeyer, Capt. James L. Tilly, Col. Thomas F. 
Van Natta, III, Lt. Gen. Albert C. Wedemeyer, Brig. Gen. Theodore F. Wes- 
sels, Capt. Logan E. Weston, Lt. Gen. Raymond A. Wheeler, and Col. Walter 
S. Wood. 

The authors have been aided by and are grateful for the opportunity to dis- 


cuss the history of the China-Burma-India theaters with General Boatner, Lt. 
Col. Dwight E. Brewer, Lt. Col. Charles F. Byars, Colonel Converse, Maj. Gen. 
William M. Creasy, Mr. John P. Davies, Jr., General Dorn, Colonel Dupuy, 
Colonel Easterbrook, General Evans, Maj. Gen. Francis W. Festing (British 
36th Division), Colonel George, General Hearn, Col. Thomas J. Heavey, 
Colonel Hill, General Hurley, General Lindsey, General George C. Marshall, 
General Merrill, Colonel Peers, General Roberts, Colonel Rockis, Col. Robert 
F. Seedlock, Colonel Spaht, Col. Joseph W. Stilwell, Jr., Colonel Stodter, Col. 
Thomas F. Taylor, Captain Tilly, General Wessels, General Wedemeyer, and 
Colonel Wood. They are of course not responsible for any errors of fact or in- 
terpretation in the pages following. 

A great deal of assistance has been received from other historians. Maj. Gen. 
S. W. Kirby, C.B., C.M.G., O.B.E., M.C., and Brigadier M. R. Roberts, D.S.O., 
of the Cabinet Office Historical Section, commented on a draft of the manu- 
script at very considerable length. Col. Allison R. Hartman and the staff of 
former Japanese officers of the Historical Section, Far East Command, pre- 
sented the authors with interpretations and corrections from the Japanese point 
of view. Dr. Henry Lee Bowen of the Air Force historians gave generously of 
his time and experience in commenting on Chapter III. 

Within the Office, Chief of Military History, the authors were greatly aided 
by Mr. Joseph Bykofsky, Transportation Section, himself a veteran of the North 
Burma Campaign. Stilwell's Command Problems like its predecessor, Stilwell's 
Mission to China, reflects much of the industrious research, judgment, and nar- 
rative skill of Mrs. Jacqueline Perry Griffin, research assistant. The section of 
Chapter II entitled "SEAC Tries To Salvage Burma Operations" was written 
by Mrs. Griffin, whose special field was the operations of Southeast Asia Com- 
mand. Only Mrs. Griffin's departure in October 1950 kept her direct contribu- 
tion to the text from being a larger one. 

Mrs. Dorothy Campbell Liles prepared the manuscript of Chapters I 
through VI in first draft, and prepared Part Two for circulation among partic- 
ipants in the campaign. Her successor, Miss Mildred Bucan, prepared the final 
draft of the manuscript for publication. Their quick perceptions, sound com- 
mon sense, and stenographic skill permitted the authors to concentrate on their 
proper tasks of research and writing. 

This volume, like its predecessor, was edited by Miss Ruth Stout, whose 
skill, industry, and understanding of authors' ways greatly improved the orig- 
inal manuscript. Miss Gay Morenus, who copy-edited Stilwell's Mission to China, 
has not only performed that exacting task most competently for the present 
volume but has prepared an extremely helpful index. 

Finally, though this volume and the others of the China-Burma-India 
series are prepared and published under the auspices of the Office, Chief of 
Military History, the reader must not assume they present an official point of 
view on wartime events in China, Burma, and India. Indeed, the authors very 


soon learned that the witnesses to and participants in these events differed 
among themselves as witnesses always do. The authors are not aware that there 
exists in the Department of Defense anything that may be called an official 
point of view toward the events described below. The point of view here pre- 
sented is that of the undersigned. Responsibility for it, and for any errors the 
reader may discover, is solely theirs. 

Washington, D. C. CHARLES F. ROMANUS 





Plans and Preparations for the North Burma Campaign : 
October-December 1943 

Chapter Page 


Combined Chiefs Order a North Burma Campaign 9 

Improving the Lines of Communications 11 

The B-29 Project Approved 15 

Chennault's Operations 18 

Chennault's Proposals for 1944 22 

Building an East China Army 26 

The Allied Command Structure in North Burma 28 

The Chinese Forces 32 

The American Force 34 

The Kachin Rangers 36 

The Battleground 37 

Planning the North Burma Campaign 39 

The Campaign Begins 45 

Summary 48 


Drafting SEAC's Proposals 49 

The United States Prepares for the SEXTANT Conference 52 

The Chinese Prepare for SEXTANT 56 

Presenting CHAMPION at Cairo 59 

Trying To Reach Agreement 63 

Over the Watershed: The Changed Attitude Toward China 67 

S til well' s Search for Guidance 71 

SEAC Tries To Salvage Burma Operations 75 

A Changing U.S. Attitude 79 

Summary 82 


Command and Administration of the Air Forces 83 

The Japanese Air Effort in Burma 85 

Support of Ground Forces 88 

Strategic Bombing 91 

Air Supply 95 

Allocation of Transport Aircraft 98 

Administrative Problems 100 

Techniques 102 

Problems of Air Supply 106 

The B-29 Command Problem 109 

Logistical Problems of the B-29's 114 

Summary 116 



The North Burma Campaign : December 1943-August 1944 

Chapter Page 


The Chiangs Visit Their Troops 122 

Yupbang Ga 124 

The Opponents Shape Their Plans 129 

Enveloping the Japanese Left Flank 131 

The Capture of Taihpa Ga 134 

Clearing the Taro Plain 136 

The Allies Reorganize for the Next Effort 138 

Logistical Support 139 

Planning To Force a Decision 142 

U.S. Infantry for the Second Phase 146 

The Operation Begins 148 

GALAHAD' s Fight at Walawbum 151 

The "Big Squeeze Play ' 155 

Summary 159 


The SEXTANT Decisions Challenged 160 

The Japanese Create More Command Problems 165 

Mountbatten and Stilwell Meet 169 

The Chief s of Staff Reject CULVERIN 171 

The Japanese Attack Forces Hump Diversions 172 

Inkangahtatvng: An Attempt That Failed 175 

The Japanese Delay the 1st Battalion, 5307th 183 

Delay at Jambu Bum 185 

The Siege of Nhpum Ga 188 

Air Supply Problem at Imphal 191 

The Chindits Go Back to Burma 196 

The Question of Myitkyina 200 

Summary 203 



The lid Division and the Drive on Kamaing 207 

The }8th Division: The Generalissimo and Stilwell 211 

Turning Tanaka s Flank 213 

The 112th Regiment's Stand at the Seton Block . 215 

Defeat of the 18th Division 218 

Stilwell and the Chindits 220 

The March to Myitkyina 223 


The First Attempts To Take the Town 229 

Japanese Build-up at Myitkyina 230 

Command Problems at Myitkyina 238 

The Attacks of Mid-June 1944 244 

Changes in Command 248 

Hacking Out Small Gains 250 

The Last Days 252 

Summary 256 


Chapter Page 


The Headquarters Reorganized 257 

SOS Problems 259 

Negotiations 260 

Clearing the Port of Calcutta 263 

Railway Problems 265 

Military Railway Service Begins Its Work 266 

Changes in Operating Procedure 267 

Operations Under Military Railway Service 269 

Physical Improvements on the Railway 271 

A British Appraisal 272 

Attempts To Use Indian River Transport 273 

Pipelines in India 274 

Supply Problems in India 276 

New Agreements on Local Procurement 277 

Housekeeping Problems 280 

Chinese Lend-Lease 281 

Medical Problems in the Rear Area 284 

Logistical Problems in China 288 

Summary 293 


Command Problems in China Theater 


"Money Is the Root of All Our Trouble" 297 

American Military Observers in North China 302 

SEAC and Stilwell Obtain Pressure on Chiang 304 

The Generalissimo Warns of Trouble 306 

The President Demands Action 309 

The Chinese Decide To Cross the Salween 312 

Chennault Renews His Warnings 314 

Operation ICHIGO 316 

East China's Defenders on the Eve of ICHIGO 320 

Initial Reactions to ICHIGO 322 

The East China Army Written Off 326 

Summary 327 


Battleground Above the Clouds: The Salween Front 331 

The Chinese Plan for the Salween Campaign 333 

The American Contribution 336 

Beginning the Offensive 340 

Pushing Through Ma-mien Pass 343 

Clearing Ta-tang-t%u Pass, 11 May-12 June 1944 345 

Securing the Huei-jen Bridge Area, 11 May-14 June 1944 349 

Driving the Japanese Rear Guards From the Shweli Valley 349 

The Southern Flank, 11 May-30 June 1944 352 

71st Army and the Fight for Lung-ling, 28 May-15 July 1944 .... 354 

Summary 360 


Chapter Page 


Stilwell's Mission Laid Aside 362 

S til we 1 1 Called to China 364 

Chennault Given 10,000 Tons 367 

The Japanese Drive Rolls on in East China 37I 

Vice-President Wallace Suggests Stilwell's Recall 374 

Stilwell Nominated for Command 379 

The Generalissimo Agrees "in Principle" 384 

The Ledo Road Project Reduced 387 

Slow Progress Across the Sal ween 389 

The Battle for Sung Shan 394 

Summary 398 


The Defense of Heng-yang 399 

Japanese Successes and Chinese . Politics 401 

The East China Crisis Grows 405 

The Hurley-Nelson Mission 413 

Washington Plans To End Stilwell's Lend-Lease Powers 418 

The Talks Begin 422 

American Proposals, 12-13 September 1944 426 

Stilwell and the Question of the Communists' Role 430 

Crisis in the East, Crisis in the West 433 

"I Do Not Seek the Job" 436 

The President Replies 439 

Summary 442 


The Note Delivered 443 

The Generalissimo s Wrath 447 

Deadlock 451 

End of the Deadlock 456 

The Generalissimo Places the Blame 460 

Waiting for the Decision 464 

The President Ends CBI Theater 468 

Conclusion 471 





INDEX 491 



No. Page 

1. Stilwell in the CBI Chain of Command: December 1943-June 1944 ... 6 

2. Organization of U.S. Forces, China, Burma and India: November 1943- 

April 1944 " . . 7 

3. Chih Hui Pu 30 

4. Japanese Organization and Dispositions: November 1943 43 

5. Tonnage Shipped From India to China by Air: 1944 112 

6. Schematic Order of Battle of the Chinese Expeditionary Force .... 333 

7. Tonnage Forwarded by USAF SOS CBI Advance Section to North Burma; 

January 1944-May 1945 390 

8. Organization of the China Expeditionary Army: 15 September 1944 . . 406 


1 . Tonnages Shipped to Burma Over Burma-Siam Railway : November 1943- 

August 1945 94 

2. Average Daily Traffic at Terminals of Principal Branch Lines of Japanese- 

Operated Railways in Burma: January 1943-August 1945 96 

3- Tonnage Delivered to Northern Combat Area Command by Air: April 

1943-March 1945 105 

4. Strength of U.S. Army Forces in the China-Burma-India Theater: Janu- 

ary-September 1944 258 

5. Fourteenth Air Force Aircraft Inventory by Type of Aircraft: March 

1943-December 1944 370 


1. India-China Communications System, October 1943-October 1944 11 

2. The Battleground 37 

3. Albacore Plan, 8 August 1943 40 

4. Entering the Hukawng Valley, October 1943 44 

5. Disposition of Forces, 1 December 1943 120 

6. Operations in Northern Hukawng Valley, January 1944 132 

7. Advance to Walawbum, 23 February-4 March 1944 144 

8. Fight at Walawbum, 4-8 March 1944 151 

9. Japanese Attack on Arakan Front, 4-7 February 1944 166 

10. Japanese Offensive on Imphal Front, March-April 1944 173 

11. Inkangahtawng, 12-23 March 1944 179 

12. Shaduzup and Nhpum Ga, 24-29 March 1944 184 

13. Mogaung Valley, 1 April-27 May 1944 207 

14. Mogaung Valley, 28 May-26 June 1944 217 

15. Advance to Myitkyina, 28 April-17 May 1944 224 

16. Myitkyina, 18 May-End of July 1944 231 

17. Situation in China, 15 March 1944 307 

18. ICHIG0 Plan 317 

19. The Sal ween Campaign, 11 May-30june 1944 330 




The Kweilin Infantry Training Center 27 

A Squad of Kachin Rangers 37 

Cairo Conference 60 

Chiang Kai-shek at Ramgarh 67 

Bombing of Bridges 93 

Kickers Prepare to Drop Supplies 98 

Packing Supplies for Airdrop 103 

Parachute-Covered Foxhole 109 

Attack on Yupbang Ga 126 

Japanese Fifteenth Army Commander and Staff 129 

General Stilwell and General Sun 135 

Drainage Culverts 140 

Pipeline Crossing a Stream 141 

General Stilwell and General Merrill 149 

A Chinese Cook 157 

Troops of Merrill's Marauders 177 

Damaged M3A3 Tanks 186 

Troops of the Chinese 22d Division 209 

A British Antiaircraft Unit 227 

Japanese Defensive Positions 232 

Allied Casualties at Myitkyina Airfield 234 

Japanese Trenches 236 

Artillery in Action at Myitkyina 246 

General Stilwell and Colonel Hunter 249 

The King George Dock Area 263 

Railroad Operations in India 268 

Supplies in Open Storage in India 282 

Bridges Across the Yellow River 324 

Gen. Wei Li-huang 334 

An American Liaison Officer 337 

American Engineers on the Burma Road 339 

Chinese Troops Cross the Salween 342 

Footbridge Across the Salween 344 

Air-Supply Drops 347 

Chinese 2d Army Artillerymen 353 

The Sung Shan Mountain Area 358 

Evacuation of Kweilin 373 

Vice-President Wallace 375 

Chinese Infantrymen Rest on Lai-feng Shan 392 

The Walled City of Teng-chung 393 

Chinese Troops on Kung Lung-po Peak 396 

Field Marshal Shunroku Hata 407 

Gen. Yasuji Okamura 407 

Chungking Conference 422 

General Stilwell Entertains at Luncheon 425 

The illustrations are from Department of 
Defense files, except those on pages 129 and 407, 
which are Japanese photographs. 





The Last Preparations 

Lt. Gen. Joseph W Stilwell's concern with command problems in China, 
Burma, and India can be traced to the Arcadia Conference in Washington, 
December 1941, when Prime Minister Winston S. Churchill and President 
Franklin D. Roosevelt proposed to Generalissimo Chiang Kai-shek that 
China's leader become Supreme Commander of a United Nations "China 
Theater." The Generalissimo agreed and, in replying, asked that a high-ranking 
U.S. Army officer be sent to China to be chief of staff of the Allied staff that the 
Generalissimo proposed to organize to help him command China Theater. The 
United States nominated Stilwell, and the Chinese agreed. This then was the 
basic command structure of China Theater, which was geographically synony- 
mous with China. The Generalissimo was Supreme Commander; all United 
Nations forces in China Theater were under him, while he in turn was answer- 
able to himself alone, in no way subject to any other officer or agency of the 
United Nations. Gen. Ho Ying-chin was Chief of Staff of the Chinese Army 
while General Stilwell was the Generalissimo's joint (Allied) chief of staff for 
China Theater. The Chinese were not willing to let Stilwell have a staff to 
help him carry out his duties as joint chief of staff to the Generalissimo, so 
Stilwell tried to perform them himself with the aid of an interpreter and a 
stenographer. And since Stilwell took this course, his American subordinates 
were often not aware of his plans for China Theater. 

Believing that the Chinese Army was not an effective fighting force, the 
War Department ordered Stilwell "to assist in improving the combat efficiency 
of the Chinese Army." In endeavoring to carry out this mission, Stilwell, after 
he arrived in China in March 1942, met with little interest or co-operation 
from the Chinese Government. An undeclared truce existed along the front, 
broken occasionally when the Japanese forces advanced to break up Chinese 
troop concentrations or to train their own troops. On such occasions, the 
Chinese fell back, then reoccupied their former positions when the Japanese 
retired. In September 1943 Stilwell told the Generalissimo that many of the 300 
divisions on the Chinese order of battle had never been in combat. About thirty 
Chinese divisions were commanded by officers whose loyalties were primarily 
to the Generalissimo; the others were loyal to local war lords or provincial 



governors. Of materiel, the Chinese said they had about 1,000,000 rifles, 83,000 
machine guns, and 7,800 trench mortars. Artillery was so widely dispersed that 
no division had enough, but there were about 1,330 cannon in China of diverse 
calibers and origins. Units were far below strength; soldiers were unpaid, poorly 
fed, and poorly clad; the diseases that accompany malnutrition and insanitary 
camps were rampant. The Chinese had not succeeded in creating a services of 
supply; consequently, troop movements were made only with the most extreme 
difficulty, for trucks and motor fuel were almost nonexistent in China and 
there was no organization for keeping up a steady flow of rations to troops on 
the march. In his memorandums and staff studies Stilwell always described the 
Chinese Army as "immobile." 

After the Japanese occupied Burma in May 1942 and destroyed the last line 
of communications between China and her Allies, Stilwell faced a problem that 
required a multiple solution if he was to carry out his orders from Gen. George 
C. Marshall, Chief of Staff, U.S. Army, "to support China." 

The solution that Stilwell proposed to the Chinese, British, and U.S. 
Governments in May and June 1942 was to form within the Chinese Army an 
elite force of full-strength, well-fed, competently led and well-trained divisions, 
the gaps in whose equipment would be made good by lend-lease aid. To bring 
artillery, small arms ammunition, shells, trucks, and spare parts into China, 
Stilwell proposed to retake all Burma and reopen the line of communications 
from Rangoon to Kunming. Once this had been done, and a powerful Chinese 
Army, supported by an adequate line of communications, had been created, 
then Stilwell (and Marshall) believed the time would be at hand for a powerful 
air offensive against Japan that would deal punishing blows to the Japanese 
homeland itself. While these proposals were being considered by the several 
governments, Stilwell in July 1942 organized the U.S. air force and service 
troops in China, Burma, and India into an American theater of operations, 
"U.S. Army Forces in China, Burma and India." 

From May 1942 to October 1943 Stilwell, with earnest support from General 
Marshall and Secretary of War Henry L. Stimson, urged his proposals on the 
Generalissimo. In Washington, Marshall and Stimson sought to enlist the 
President's aid. The basic factor controlling the support that StilwelPs American 
superiors would extend to his efforts to assist China was the American decision 
to make the principal U.S. effort in the Atlantic area. Stilwell's theater, there- 
fore, with the exception of transport aircraft, received little in the way of 
supplies and manpower from the United States; to a great degree Stilwell was 
left to carry out his mission with what resources he could conjure up in China 
and India. 

Some progress was made toward creating a better Chinese Army. Detailed 
plans were made to assemble thirty reorganized and re-equipped Chinese divi- 
sions in Yunnan, and thirty more in east China. An infantry and artillery train- 



ing program in Yunnan was begun in March 1943. The nucleus of an American 
liaison staff was set up for the Yunnan divisions, or Y-Force as they came to be 
called. But every step of the way Stilwell and his subordinates had to cope with 
what they believed to be apathy and indifference on the part of the Chinese. 

After a brief period in the fall of 1942, when it appeared that the Generalis- 
simo might be actively interested in reforming a major portion of the Chinese 
Army and joining in an offensive to retake Burma, he stated that operations in 
Burma in March 1943 could not be undertaken and invited attention to what 
might be done in China by a small, effective air force. 

The commander of the Fourteenth U.S. Air Force in China, Maj. Gen. 
Claire L. Chennault, believed that Stilwell, by concentrating on reform of the 
Chinese Army and reopening of the ground lines of communications to China, 
was compromising a great opportunity to deal heavy and immediate blows 
against the Japanese by air. Chafing at the restrictions placed on him by the 
small amount of supplies then being flown into China from India, General 
Chennault in October 1942 finally put his case directly before the President. He 
told Roosevelt that with 105 fighters, 35 medium bombers, and 12 heavy 
bombers, he could open the way for the defeat of Japan. By March 1943 Roose- 
velt gave unmistakable indication that he supported Chennault rather than 

At the Trident Conference in Washington during May 1943, the divergent 
trends in U.S. policy finally came into the open. The Generalissimo as Supreme 
Commander, China Theater, asked the President to give Chennault the logis- 
tical support that officer said would suffice. The President agreed, and Chen- 
nault received first priority on supplies flown into China. The War Department 
advised against giving first priority to Chennault, because it feared that if the 
Japanese were provoked by Chennault's air offensive they would sweep over the 
east China bases from which the Fourteenth Air Force operated. The Generalis- 
simo, however, gave Roosevelt his personal assurance that the existing Chinese 
forces could defend the east China airfields. 

The second major development at the Trident Conference was that the 
Combined Chiefs of Staff decided not to attempt the reconquest of all Burma 
in the near future but rather to reoccupy north Burma only. Such a limited 
campaign meant that in lieu of what might come through Rangoon and over 
the old Burma Road, supplies for the Chinese Army and Chennault's air force 
would be confined to what could be brought across north Burma— by air until 
the campaign succeeded, by air, truck, and pipeline thereafter. After the 
Trident Conference it speedily appeared that Allied resources in China- 
Burma-India (CBI) were insufficient to support both Chennault's air offensive 
and preparations to retake north Burma. 

The need for a new approach to the problems of war in Asia and for 
adequate logistical support of Allied operations there was by then obvious. The 
solution put forward by the Prime Minister in June 1943 and agreed to by 



Chart 1 — Stilwell in the CBI Chain of Command : 
December 1943-June 1944 








tZcttnlh Am 




f OlM+fffltd Army 

Chinktt Army 

■n Indiii 


ml Oiitl* 




Dtpviy CG 

China Tn«H*f 

Sector, AAF 

Chief al Staff 
CWo Tfiiaicr 


. T*„ih Af 


FeuflMnlh AF 

& Combat Comd 

• »»*•• H*orJqfiortrtti . NC AC, undtr If mpaicry op*'3'" 
— * — Lioiion 
™ Common*!, 

3> control, 90 June 1944 

American authority was the creation of Southeast Asia Command (SEAC) 
under Vice Adm. Lord Louis Mountbatten as Supreme Allied Commander with 
Stilwell acting as his deputy. An ambitious program of engineering projects was 
begun. When Mountbatten visited China in October 1943 to present his new 
command to the Generalissimo, the latter requested that Stilwell be recalled. 
Though Stilwell and the Generalissimo were not on good terms, though the 
latter usually ignored Stilwell both personally and officially, this request came 
immediately after a very brief period in which the Generalissimo had shown a 
renewed interest in StilwelPs proposals for reform of the Chinese Army, and 
therefore it surprised and angered Stilwell. 

The Generalissimo's request climaxed about twenty months in which 
Stilwell and his subordinates had sought to improve the combat efficiency of 
the Chinese Army. Armies do not spring from the ground fully armed and 
trained; months of time are required for the task of preparing them. The cam- 
paigning season was about to begin in the fall of 1943, and the months that 
should have been used for China's preparation were sped. On the eve of its 
greatest test the Chinese Army in China was little better than it had ever been, 
and Stilwell was convinced there was little more he personally could do in 
China. Henceforth, he gave most of his time and attention to command 
problems in India and Burma. ( Charts 1 anc^T)[ 

But the diplomatic aspect of Stilwell's presence in Asia remained. His 

Chart 2 — Organization of U. S. Forces, China, Burma and India: November 1943-April 




Fwd Ech 


Rr Ech' 
Htw Delhi 
DColS (f*m,> 


Indio^bWo Sedai 

Trl-g Comd 

Fourteenth AF 


ASC h 

Command channel. 
' i Operational conbol. 
' Sm n, k 

Base Sec 1 




CT * CC 


BoteStt 1 


Ad* See 5 


Baie Sec 3 


Ae\ Sec 3* 

Adv Sm 4 4 

Kw* i I in 
(Byroad* ) 

Bombay Port of 








Combat Tit* 


■Effeetiv* T April 1944, the title Rear Ethtron woi discontinued. Theat* headquarters moved to New DeJhi, India. 
'Headquarter moved to Calcutta on IS April 1 944, 
* Discontinued December 1943. 
Inactivated 31 January 1944; consolidated into Advance Section No. 1, headquarteri at Kunming. 

'Effective 31 December 1943, Bombay Part of Debarkation e^tabliihed oi an exempted station under Commanding General, Tronjportalion Service, SOS, CBI. 
' Established 1 January 1944. 

■Discontinued 31 January 1944- Headquarters, NCAC established at Ledo an 1 February 1944. 
"Chinese Trainrnj and Combat Command, a troop pool providing officers for the or 9 oniio1ioni indicated. 




superiors, possibly fearing the diplomatic consequences if the Chinese learned 
that the U.S. Government no longer attached importance to improving the 
combat efficiency of the Chinese Army, never formally altered that as his pri- 
mary mission. Instead, they tacitly acquiesced in his conclusion that there was 
little more he personally could do in China to carry it out, and gave him a 
whole series of added tasks which, as the Army Chief of Staff later admitted to 
Stilwell, added up to what Marshall called a "paramount mission"— one that 
interfered with, and bore no relation to, his existing primary mission. 

Viewed as a whole, the war gave the United Nations grounds for confidence, 
though much hard and bitter fighting lay ahead. In the Pacific, American and 
Australian forces were battering through the outer Japanese defenses and by 
2 November 1943 were halfway up the ladder of the Solomon Islands. Once 
the island barrier of the Southwest Pacific was broken, Allied task forces could 
range more freely among the Japanese island positions. In New Guinea, the 
Allied advance was almost ready to turn the barrier represented by New Britain 
Island. But New Britain was many, many miles from Tokyo, and at times it 
may have seemed that the twenty-three months since 7 December 1941 had 
done little to bring the Allies closer to Japan. The most immediately hopeful 
augury lay in that night and day the dockyard crews at Pearl Harbor were put- 
ting oil and water, shells and fuel, into spanking-new aircraft carriers, destroyers, 
and cruisers, on landing craft and transports; that Infantry and Marine officers 
were intently studying maps of the central Pacific. Occupation of key points in 
the Gilbert Islands, Operation Galvanic, was imminent. 

In the Mediterranean, Benito Mussolini, the creator of fascism, had been 
ignominiously deposed on 25 July 1943, and a little more than a month later 
Italy surrendered to the Allies. A powerful Allied force then landed at Salerno 
(9 September). Eight days later southern Italy was firmly in Allied hands. 

In Russia, the Germans were retreating. In the summer of 1941 and again 
in that of 1942 the Germans had scored flashy victories without being able to 
strike a decisive blow. On 5 July 1943, they opened their annual summer offen- 
sive on the central front in Russia. In ten days the offensive had been halted 
and the Red Army seized an initiative it never thereafter lost. By November 
1943 Russia could begin to look forward to the day when the last German 
would be driven from Russian soil; remembering the Russo-German Pact of 
1939, the British Commonwealth and the United States might wonder what 
form Soviet policy would take in that event. 

In northwestern Europe, there was as yet no action on the ground, but 
night and day the bombing fleets flew from England to Germany, cascaded out 
their tons of fire and steel, and returned less what toll the defense took. At sea 
the submarine menace, which had so hobbled Allied strategy in 1941, 1942, and 
early 1943, was under control. So men and materiel were accumulating in Eng- 
land, and Anglo-American capabilities for a cross-Channel assault were steadily 



Of the three enemy states— Germany, Italy, and Japan— only Italy had as 
yet been invaded, and Germany and Japan still sheltered behind their outer 
defenses. In the case of Germany, the bombers were flying over the defenses, 
but in the case of Japan, systematic bombing of the Japanese homeland was as 
yet only a project for the future. China's geographic position, China's assumed 
resources in manpower, might still play a great, perhaps a decisive, role in the 
forthcoming attacks on Japan. How these Chinese resources could be brought 
to bear and the extent of British and American dependence on them for victory 
in the Pacific were among the major problems requiring solution in October 

Combined Chiefs Order a North Burma Campaign 

Stilwell began concentrating on his command problems about 23 October 
1943, when he and his immediate superior for operations in Burma, Admiral 
Lord Louis Mountbatten, arrived in India after conferences with the Generalis- 
simo in Chungking. 1 Stilwell's first task was to aid Mountbatten in the prepara- 
tion of a plan for operations in Burma to carry out the directives of the Trident 
and Quadrant (Quebec, August 1943) Conferences. 2 With the lifting of the 
monsoon rains in Burma, the campaign season was at hand, and the time had 
come to break the blockade of China in the way prescribed by the Combined 
Chiefs of Staff: 

1. to carry out operations for the capture of Upper Burma in order to improve the air 
route and establish overland communications with China. Target date: mid-February, 1944. 

2. to continue to build up and increase the air routes and air supplies of China, and the 
development of air facilities with a view to: 

a. Keeping China in the war. 

b. Intensifying operations against the Japanese. 

c. Maintaining increased U.S. and Chinese Air Forces in China. 

d. Equipping Chinese ground forces. 3 

The Myitkyina (pronounced myi-che-na)-Mogaung area of north Burma 
acquired great importance because of this directive. In the first place, its geo- 
graphic position at the southern tip of the hump of mountains over which the 
transports flew from India to China meant that its capture would greatly 
improve the air route to China. As long as the Japanese held the Myitkyina 
airstrips, the threat of their fighters forced the U.S. aircraft to fly far to the 
north, then to swing south to the Kunming air terminals. This increased fuel 

1 Report and Supplement for the Combined Chiefs of Staff" by the Supreme Allied Commander, 
South-East Asia, 1943-1946, Vice Admiral Viscount Mountbatten of Burma, New Delhi, India, 
July 30, 1947 (hereafter, Mountbatten Report), Pt. B, par. 8. 

2 Charles F. Romanus and Riley Sunderland, Stilwell's Miss icmjn Chi na, UNITED STATES 
ARMY IN WORLD WAR II (Washington, 1953 ^IChs. fxl anjxl 

' History of the China-Burma-India Theater, 21 May 1942-25 October 1944 (hereafter, 
History of CBI), p. 161. OPD 314.7 CTO, A47-30 jT5ee Bibliographical Mote.) | 



consumption and cut the pay load. The air route itself was narrow, and its 
saturation with transports sometime in the near future was predicted. With 
Myitkyina in Allied possession the transports would be able to use a much 
wider, lower air route. 4 

Secondly, since fall 1942 the U.S. engineers had been building a road south 
from Ledo, Assam, which was intended to cross north Burma and ultimately 
link with the old Burma Road. The Hukawng and Mogaung valleys, down 
which the Ledo Road was being constructed, enter the Irrawaddy valley, which 
is the most habitable part of north and central Burma, within a few miles of the 
Myitkyina-Mogaung area. Both towns are on the rail and road net of prewar 
Burma, so when the Ledo Road reached them the\engineering problem would 
become one of improving existing facilities rather than constructing new ones 
in the virgin wilderness. Therefore, taking the Myitkyina-Mogaung area was 
the prerequisite to completing the Ledo Road and opening a ground line of 
communications, with an all-weather road and a gasoline pipeline, to China. 

The purpose of the projected ground line of communications to China was 
not to supply the Chinese armies or sustain the Chinese economy; a road and 
a pipeline cannot support 300 divisions in combat or sustain the life of several 
hundred millions of people. A two-way, all-weather road to China, which was 
then contemplated, would, the Army Service Forces believed in the fall of 1943, 
permit the ultimate delivery of 65,000 tons a month to Kunming. Vehicles, 
artillery, and small arms ammunition from this tonnage would enable Stilwell 
to fill the gaps in the equipment of such Chinese divisions as the Generalissimo 
might permit him to train and bring to full strength/ Once revitalized, these 
divisions might, if the Generalissimo concurred, seize a port on the Chinese 
coast and secure airfields from which Japan could be systematically and heavily 

Normally, planning precedes logistical preparation, and logistical prepara- 
tion, fighting. One of the noteworthy aspects of the North Burma Campaign 
of 1943-44 is that the logistical preparations, the planning, and the fighting 
proceeded simultaneously. The troops moved forward before the commanders 
agreed on their plans, and the logistical preparations were months in being 

4 (1) See remarks of Col Francis Hill. Min. Washingto n Conf, 9 Feb 44. ABC 384 (Burma), 
S-2 5-42 Se c 5, A48-224 . |(See Bibliographical Note.)| (2) See Stilwell' s Mission to China re 
|TRiDENii and | Quadrant. ^ 
""'"TT) The principal Army Service Forces and CBI Theater headquarters studies of the projected 
capacity and requirements of the Ledo Road are discussed in a manuscript by Joseph Bykofsky, 
The History of Transportation Service in China, Burma, and India in World War II. (Hereafter, 
Bykofsky MS.) (See Bibliographical Note.) (2) The principal staff studies are: 1. Ltr, Col Frank 
Milani, AG, Rr Ech, USAF CBI, to Marshall, 31 Jan 44, sub: Project TIG 1-C. Folder, AG (537) 
Transportation Sec, SOS USFCT, KCRC 2. Rpt, OCofT ASF, 10 Feb 44, sub: Proposed Motor 
Transport Service, CBI. Hist Br, QCofT. 3. History of Services of Supply, China, India, Burma 
Theater, 28 February 1942-24 Oct ober 1944 (hereafter, SOS in CBI), App. 24, Sec. 2, Pt. 1. 
OCMH. (See lbibliographical ^^^Jor these references.) 



Improving the Lines of Communications 

The Quadrant Conference, held at Quebec in August 1943, resulted in an 
ambitious list of engineering projects for In dia and B urma to increase the 
movement of supplies from Calcutta to China \{M.ap 1)\ 

1. An increase in the amount of air cargo being flown to China by the Air 
Transport Command (ATC) to 20,000 tons a month by mid-1944 

2. A road from India to China (the Ledo Road) with an initial (January 
1945) capacity of 30,000 tons per month 

3. A gasoline pipeline from Assam via Fort Hertz in northernmost Burma 
to Kunming 

4. A thin-walled 6-inch pipeline from Calcutta to Assam Province to supply 
the Air Transport Command airfields there 

5. A thin- walled 6-inch pipeline to China 

6. An American-operated barge line on the Brahmaputra River to bring 
supplies forward from the great port of Calcutta to the Allied bases in Assam 

7. Improvement of the Bengal and Assam Railway 6 

Basic to this program was the realization that many of the difficulties 
hobbling the Allied effort in Asia sprang from the fact that the line of com- 
munications from Calcutta to the airfields and storage depots of Assam could 
not bring supplies forward in sufficient quantity to support a major effort in 
the air and another on the ground. To General Headquarters (India), which 
was responsible for logistical support of the British and India forces on the 
border of Burma, Lt. Gen. William J. Slim, commanding the Fourteenth Army, 
wrote on 30 October 1943: 

The supply situation as regards certain commodities in the Army area is so serious that 
I consider it will affect active operations and should, therefore, be brought to the attention of 
the Commander-in-Chief [Gen. Sir Claude J. E. Auchinleck] and the General Officer 
Commanding-in-Chief 11 Army Group [Gen. Sir George Giffard]. ... In general terms, 
instead of holding a tonnage of 65,000 tons which is the target for the supply depot in the 
area, only some 47,000 tons were held on 26 Oct 43, thereby giving an over-all [Slim's italics] 
deficiency of 27%. . . . 

Taking up the supply situation area by area, General Slim noted that in the 
Arakan district of Burma there was no hay for animal transport or clarified 
butter in forward areas. In the area forward of the Manipur Road British troops 
had no meat, nor were authorized substitutes available. For Indian troops in 
the same vicinity there was no meat, no milk, no butter, the latter two staple 
items of the Indian diet. Moreover, when tinned foods did come forward, often 
50 percent was spoiled. Therefore, "in 4 Corps area the medical authorities have 
already reported that troops are suffering from malnutrition." 7 Admiral 

6 Rpt, Col Frederick S. Strong, Jr., Chief Engr, SOS CBI, to Lt Gen Brehon B. Somervell, CG, 
ASF, 13 Nov 43. Somervell File, Vol IV, Hq ASF, Theaters of Opns, CBI 1944, A46-257. (Here- 
after, Somervell File .) 

7 Ltr, 3872/ 17/Q, Slim to GHQ (India), 30 Oct 43. SEAC War Diary. (See Bibliographical 
Note.) 1 



Rail ROADS : 

— Broad gauge . cO'lisle track 

■ — Broad &AUG£ , SINGLE IRAC* 

' 1- meter gauge , single track 

— Burma roao 

== = = = = = = Other HCAOS 

► Barge route 

t Pipeline in operation 
Pipeline projected 

^ . AiRLif T TO China 

Q Major airfields 







Mountbatten was keenly aware of the intimate connection between the Assam 
line of communications and the prospects of victory and placed improvement 
of the Bengal and Assam Railway very high on his agenda. 8 

Vital though it was, the Bengal and Assam Railway was but part of the 
Assam line of communications. The major factors lessening the efficiency of the 
line of communications were (1) the congested port of Calcutta, through which 
supplies for Assam and China entered India; (2) the inefficient rail lines and 
barge lines which moved them forward; (3) the civilian agencies of the Gov- 
ernment of India which controlled the flow of goods and personnel over the 
line of communications. In point of time, the railway was considered first, but 
ultimately each of the three factors above had to be appropriately handled, and 
gasoline pipelines constructed, before a satisfactory solution could be reached. 

As a depressing backdrop to the war effort in Bengal, famine racked that 
great province, in which lay the all-important harbor of Calcutta. The rice crop 
of 1943 had failed to meet the demands on it; the little skiffs and luggers that 
plied the intricate Bengal waterways in peacetime had been destroyed by British 
authority in 1942 for fear of a Japanese invasion, and the Japanese occupation 
of Burma had effectively cut off a major source of Indian rice. Before the famine 
ended, more than 1,000,000 Bengalis died. The famine held full and horrible 
sway in the fall of 1943; it was a heavy added burden on the already strained 
provincial and central authorities, whose reactions to the needs of war along 
India's eastern frontier must be judged against the emergent needs of India 

On 23 October when Lt. Gen. Brehon B. Somervell visited India, the 
American supply expert volunteered to provide U.S. railway troops to assume 
some of the burdens of operating the Bengal and Assam Railway. His offer 
began a discussion between interested agencies of the Government of India and 
SEAC. As a result of the initial exchanges the director of transportation of 
General Headquarters (India) and the general manager of the railway made a 
rapid survey of the problems that would be created if Americans helped to 
operate key sections of the line. Their report agreed to U.S. assistance. The 
Government of India was guided accordingly, accepting it on 6 November, and 
on 10 November CBI Theater told the War Department that accord had 
reached the point at which details were being discussed. 9 

Anticipating such success, General Somervell, Maj. Gen. Charles P. Gross, 
Chief of Transportation, Maj. Gen. Raymond A. Wheeler, Commanding Gen- 
eral, Services of Supply, CBI, and their staffs had earlier agreed on the nature of 
the troops required. Wheeler alerted the War Department when the Govern- 
ment of India's acquiescence appeared reasonably certain, and so the actual 
movement followed quickly on India's agreement. Orders were issued on 
16 November to the 705th Railway Grand Division, the 758th Railway Shop 

8 Extract, SAC's Personal Diary, 30 Oct 43. SEAC War Diary. 

9 SOS in CBI, pp. 53-56. 



Battalion, and the 721st, 725th, 726th, 745th, and 748th Railway Operating 
Battalions. Each battalion was less its maintenance of way company, for the use 
of Indian labor was contemplated. Col. John A. Appleton, former general 
manager of the New York zone of the Pennsylvania Railroad, arrived in India 
on 16 November as a member of the Appleton-Inglis Railway Mission. He 
was known to be the War Department's choice for the post of director of the 
proposed Military Railway Service. 10 

Therefore, as of mid-November 1943 it was agreed by all concerned that 
American troops would assist in operating certain key sections of the Bengal 
and Assam Railway. Exact definition of the sections and the conditions of 
operation were being negotiated, and the troop movement was under way. 

A factor in these negotiations was the report submitted by Col. Paul F. 
Yount, of the Transportation Corps. Sent by Somervell from Iran, where he had 
succeeded in increasing the amount of tonnage carried by the Iranian rail system 
to 500 percent above prewar standards, Yount made a rapid reconnaissance of 
the Bengal and Assam Railway. His report of 10 November strongly suggested 
that more efficient and more vigorous operating methods, rather than the 
tedious processes of rebuilding the railway, would bring a sharp and immediate 
increase in the tonnage carried. 11 The greatest deficiency, he believed, was lack 
of an aggressive, adequate supervisory staff between management and actual 
operating personnel. 

Mid-November saw major personnel changes among the men guiding the 
logistic effort. General Wheeler, chosen by Admiral Mountbatten to "push" 
improvements in the transportation system, became principal administrative 
officer of SEAC on 15 November. Lt. Gen. Sir Wilfred Lindsell, offered for the 
same post by Gen. Sir Alan Brooke, Chief of the Imperial General Staff, became 
instead principal administrative officer for Gen. Sir Claude J. E. Auchinleck's 
India Command. General Wheeler was succeeded in command of the SOS, 
CBI, by Maj. Gen. William E. R. Covell. 12 

At the extreme northeastern end of the vital railway line, construction of 
the Ledo Road was resumed as the end of the rains and the gradual drying of 
the ground permitted work to begin. The 849th and 1883d Engineer Aviation 
Battalions and the 382d Engineer Battalion (Separate) had arrived in September 
to reinforce the road builders. On 17 October Col. Lewis A. Pick assumed 
command of the road project as commanding officer of Base Section No. 3. 
Before coming to CBI, Colonel Pick had been division engineer of the Missouri 
River Division, Omaha, Nebraska, and had drawn up the Pick Plan of Missouri 
River flood control. Three more units arrived in November: the 45th Quarter- 

10 SOS in CBI, pp. 60-62. 

11 (1) Ltr, Yount to Wheeler, 10 Nov 43, sub: Capacity of Assam LOC; Ltr, Somervell to 
Wheeler, 25 Oct 43. Somervell File, Vol IV, CBI 1944. (2) For Yount's work in Iran, see T. H. 
Vail Motter, The Persian Corridor and Aid to Russia. UNITED STATES ARMY IN WORLD 
WAR II (Washington, 1952). 

12 (1) SOS in CBI, pp. 29-30. (2) Mountbatten Report, Pt. A, par. 39. (3) Ltr, Somervell 
to Wheeler, 27 Oct 43. Somervell File, Vol IV, CBI 1944. 



master Regiment, the 209th Combat Engineer Battalion, and the 1905th 
Engineer Aviation Battalion. 

From the end of October to the end of November, the lead bulldozer 
advanced over twenty-two miles, to mile 82.35 from Ledo. To speed construc- 
tion, three bulldozers left the village of Nawngyang on 28 August to make 
their way over a trail, once used by the refugees fleeing Burma in 1942, to the 
village of Namlip Sakan, which lay on the trace of the Ledo Road. Progress 
was extremely difficult, but the dozer crews brought their machines through 
the monsoon rains to their goal on 5 October 1943. From Namlip Sakan they 
began cutting trail in both directions, back toward Ledo and forward into 
Burma. With dry weather from late October, actual road construction, as 
distinguished from clearing the way, made good progress. Anticipating that 
the road would soon be through the mountain barrier into the Hukawng Valley, 
Colonel Pick sent nineteen men ahead to the village of Shingbwiyang (mile 103 
from Ledo on the southern end of the Patkai Range) to establish a supply 
depot for the first truck convoy. At Shingbwiyang Pick's little supply detach- 
ment was about one third of the way to Myitkyina. 13 

Surveying the engineering projects on 1 November, the SOS chief engineer 
told General Somervell that better progress would be made with the Ledo Road 
once the men and machinery on requisition were deployed south of Shing- 
bwiyang, but he warned, "Over-optimism on road construction, if more 
evidence is needed than that furnished at Ledo, and on road capacity, would not 
be indulged in if the true lessons of the Alcan and Pan American Highways 
were generally known." 14 Somervell believed that the Ledo Road would reach 
Pao-shan, China, by 1 November 1944, an estimate which was, of course, an 
engineering one since no one could foresee just when the Japanese would be 
driven from the trace of the Ledo Road. 15 

In November the construction of pipelines that would ultimately stretch 
from India to China began. Shortly after the Quadrant Conference theater 
authorities had decided to abandon the Fort Hertz pipeline. Proponents of this 
route had long argued that constructing a pipeline by this short route might be 
enough when added to the Hump airlift to support the Fourteenth Air Force 
on an ample scale. The mountainous terrain the Fort Hertz line would have to 
cross was a powerful argument against the project. 16 Four other reasons influ- 
enced abandoning the project for which such hopes had once been entertained: 
(1) it was not possible to bring men and materiel to CBI Theater as early as 
they would be needed; (2) the Assam line of communications was congested 
and would not permit bringing materiel forward as rapidly as needed; (3) it 

13 (1) For background on the CBI engineering projects, see \Sti/weI/'s Mission to China.\ {2) 
SOS in CBI, p p. 441- 43. *" - i 1 

14 Rpt cite d n. 6.1 

15 Memo, Somervell for JCS, 24 Nov 43, sub: Progress Rpt on Bengal- Assam LOC. Somer- 
vell File, Vol IV, Pt I, CBI 1944. 

16 Ltr, Col Hill to Maj Gen Orlando Ward, Chief, Mil History, 2 Sep 52. OCMH. 



was feared that unavailable heavy-walled pipe would be needed for the line; 
and finally, (4) a parallel road would be necessary to maintain the line. 

Instead of the Fort Hertz pipeline, a thin- walled 4-inch pipeline would be 
built parallel to the Ledo Road. Thin- walled pipe was air transportable, and so 
the line could be built from both ends toward a meeting somewhere in Burma. 
As soon as possible a second line was to be laid from India, using heavier pipe 
where advisable, so that ultimately there would be two pipelines from India to 
China, one carrying aviation gas for the ATC and the Fourteenth Air Force and 
one carrying truck fuel. Construction began in November. 17 

The 6-inch pipelines from tanker terminals in Bengal to Assam were 
properly part of the Assam line of communications; their building presented no 
particular engineering problem. Construction of the line by eight engineer 
petroleum distribution companies assisted by 7,000 Indian laborers was con- 
templated. As 1943 ended, final preparations were being made to begin the 
work in January 1944. 18 

The B-29 Project Approved 

A new burden was added to the already overstrained China-Burma-India 
logistical structure in fall 1943, the B-29's of the XX Bomber Command. In 
August 1943 the Army Air Forces (AAF) planners had suggested bombing 
Japan into submission with an enormous force of heavy bombers based in 
China and supported by swarms of transports flying from Indian bases. 19 Asked 
to comment, Maj. Gen. George E. Stratemeyer, Headquarters, U.S. Army Air 
Forces, India-Burma Sector, pointed out that in order to support the AAF plan 
the port of Calcutta would require tremendous expansion, communications on 
a grand scale would have to be installed, north Burma would have to be cleared 
by 1 July 1944, and a variety of other Herculean projects would have to be 
successfully completed by that early date. As a substitute plan Stratemeyer's 
staff offered Twilight. 20 

Twilight assumed that north Burma would be free of Japanese by 1 July 
1944 so that the Ledo Road might speedily go through to China, and further 
that the Chinese and Fourteenth Air Forces could by 1 November 1944 be made 
strong enough to hold the east China airfields against Japanese attack. Twi- 
light's original contribution to the development of strategy in China-Burma- 
India was its suggestion that the B-29's be based in the Calcutta area and staged 
through Chinese bases to attack Japan. Stratemeyer's planners believed that 
412 transports could support 280 B-29's by hauling their supplies to the Chinese 
bases, and that Twilight would destroy Japan's ability to resist. 

17 (1) Rpt cited n. 6. (2) SOS in CBJ, p. 108. 

18 SOS in CBI, pp. 111-13. j—. 

19 Stilwell's Mission to China^Cku)^ 

20 CM-IN 9027, Stratemeyer to Gen Henry H. Arnold, 11 Sep 43. 



Examining Twilight, the Operations Division of the War Department 
remarked that the concept of staging India-based bombers through China 
offered the only hope of bombing Japan from Chinese bases in March and April 
1944 without major interference with other operations in CBI and with a high 
degree of security from a Japanese ground reaction. The Operations Division 
understood that Cheng-tu in west China would be the B-29 China base. Con- 
sistent with the views it had expressed when the President was approving 
Chennault's plans in May 1943, the Operations Division feared that a major 
Japanese reaction would overrun the Kweilin-Changsha Fourteenth Air Force 
bases in east China if the B-29's were based there. In the light of the foregoing 
factors, the China-India Section of the Theater Group, Operations Division, 
and the Strategy and Policy Group, Operations Division, both recommended 
that the B-29 project for CBI be pushed to completion with bases at Cheng-tu 
and Calcutta. 21 

Twilight's conception of the B-29's had been that of a club, bludgeoning 
Japan into submission. In November 1943 the idea grew in Washington that 
the air weapons could be used like a stiletto, striking the vital spots with a few 
skilled blows. The Japanese coke ovens seemed to offer a wonderful target. If 
they were destroyed, surely the Japanese steel industry would be crippled. In 
the fall of 1943 the Foreign Economic Administration, the Committee of Oper- 
ations Analysts, and the Office of Strategic Services (OSS) agreed in believing 
that the Japanese steel industry was producing to the limit of its capacity, 
which the analysts set at about 13,690,000 tons annually, an accurate estimate, 
for 13,970,000 tons was correct. However, they erred in believing Japan's steel 
industry was producing at capacity, for it was not, by 5,138,000 tons. Shortage 
of raw materials, not shortage of plant, was hobbling the Japanese. Therefore, 
attacks on steel plant facilities had to destroy 5,000,000 tons capacity before 
they cut current Japanese steel production. 

Fourteenth Air Force attacks on Yangtze River iron-ore carriers were 
directly cutting Japanese steel production, but this was not realized at the time. 
The highly publicized attack on Japanese ocean shipping, for which such claims 
had been made and such hopes entertained, was relatively minor in its effects, 
while the routine missions against river shipping were hitting home. They were 
mainly responsible for cutting down Japanese imports of iron ore from 2,200,000 
tons in the first half of 1943 to 1,150,000 tons in the first half of 1944. But this 
was not appreciated in Washington or in China. 22 

When the proposal to base the B-29's in India, stage them through China, 

21 Memo, Col Thomas S. Timberman, Chief, Asiatic Sec OPD, for ACofS, OPD, 2 Nov 43, sub: 
Bomber Offensive from China; Memo, Col Frank N. Roberts, Chief, Strategy and Policy Gp, OPD, 
for ACofS, OPD, 30 Oct 43, sub: Bomber Offensive from China. Case 192, OPD 381 CTO, 

22 U.S. Strategic Bombing Survey [USSBS], The Effects of Strategic Bombing on Japan's War 
Economy (Washington, 1946), pp. 44-45, 76-77, App. Tables C-16, C-17. 



and bomb Japan's coking facilities was laid before the President, he accepted it 
with enthusiasm. Approving it in early November 1943, he at once sought the 
co-operation of the British Commonwealth and China. Telling Churchill that 
Japan's steel industry was strained to the limit, the President stated that the 
B-29's could destroy half of the coke ovens supporting Japan's steel production 
and so partially cripple the Japanese. Churchill was asked to give all possible 
aid in building four airfields in the Calcutta area. To the Generalissimo, Roose- 
velt sent word that five airfields with limited facilities would be needed in the 
Cheng-tu area by 1 March 1944. He offered to give the Chinese necessary engi- 
neering supervision, supplemented by lend-lease funds if the Chinese would 
supply labor and materials. 23 

The President's initiative, regarded as "approval in principle," started the 
wheels turning, though passage of the concept through the Joint Chiefs of Staff 
( JCS) and Combined Chiefs of Staff (CCS) machinery was still to come. In 
anticipation of JCS and CCS action, engineer aviation battalions and dump 
truck companies which would be needed to build the Indian airfields were 
alerted for shipment. Their movement orders were issued on 14 November, and 
the first project personnel arrived in India on 24 November 1943. Twilight 
was reduced in scale by the provision that the B-29's would carry their own 
supplies to China, thus eliminating the 400 transport aircraft. It was intended 
that they should not be a burden on the Hump airlift. This then was the final 
edition of the plan to defeat Japan that the AAF had offered at Quadrant: 
Matterhorn— a self-supporting task force of B-29's, based on India, staged 
through China, was to cripple Japan's steel industry. 24 

Learning of the B-29 project, Stilwell wrote: "FDR has undercut me again. 
Told Peanut [the Generalissimo] all about Twilight, so I can't bargain on 
that." 25 

The reference was to Stilwell's and the War Department's consistent view 
that any American initiative in CBI should adopt a quid pro quo or bargaining 
technique. The President's military advisers had urged that American aid be 
given to the Chinese only to the degree that they agreed to help themselves. 
In March 1943 the President explicitly rejected the bargaining approach to 
China's problems. Stilwell believed that the President's rejection greatly handi- 
capped him in dealing with the Chinese, and that it was one of the reasons he 
could not persuade the Generalissimo to reform his Army. 26 

» (1) Rad, Air Ministry to Armindia, 12 Nov 43. SEAC War Diary. (2) Rad WAR 3815, 
Marshall to Stilwell and Stratemeyer, 11 Nov 43. Item 1230, Bk 4, JWS Personal File. (See 
Bibliographical Note.) 

24 Memo, Arnold for Marshall, 12 Nov 43, sub: Early Sustained Bombing of Japan; Memo, Brig 
Gen John E. Hull, Chief, Theater Gp OPD, for DCofS, 14 Nov 43, sub: Engr Units for CBI 
Theater. Case 192, OPD 381 Security, A47-30. (2) History of CBI, Sec. II, Ch. Vie, (3) Min, 
CCS 128th Mtg, 23 Nov 43. 

25 Stilwell Black Book, 12 Nov 43. (See l Bibliographical Note! 

26 For a fuller discussion of the question of bargaining with the Chinese, see |?777u^77T^Tw7off" 

to China. 



Chennault's Operations 

The autumn months of 1943 found the Fourteenth Air Force behind the 
schedule set for it by the Chennault Plan of May 1943, because the plan had 
been put into operation before its logistical requirements were on hand. Gen- 
eral Chennault had proposed to gain air superiority in China in August 1943, 
then to attack Japanese ocean shipping, and in the last phase to use B-24's for 
strategic bombing of Formosa, the Shanghai-Nanking-Hankow triangle, and 
Japan itself. To do this, which he thought would force the Japanese in six 
months to begin abandoning their outlying Pacific holdings, he requested 150 
fighters, 48 B-25's, and 35 B-24's, plus photo reconnaissance craft. For supply, 
he asked 4,790 tons per month from July to September, and 7,129 tons monthly 
thereafter. At Trident, Washington, May 1943, the President gave him first 
priority on Hump tonnage for a limited time. 27 

Events did not go as Chennault had hoped. Reinforcements were slow in 
reaching him. Hump tonnage deliveries were not in the quantities contem- 
plated at Trident. The Japanese reacted vigorously, and flying weather was 
often bad. Chennault felt profoundly discouraged as his long-sought oppor- 
tunity seemed to be slipping from him. 

Then in September 1943 the picture brightened. Hump tonnage (all car- 
riers) improved from the 5,674 tons of August to 6,719 tons in September and 
rose to 8,632 tons in October. From these monthly deliveries the Fourteenth 
Air Force received 3,038 tons, 4,575 tons, and 4,225 tons, respectively. Rein- 
forcements arrived for the Fourteenth Air Force, and it swiftly began to change 
from the semiguerrilla force of earlier days to a powerful weapon, whose subor- 
dinate commands had area assignments. Arrival of the 25 th and 26th Fighter 
Squadrons in October reunited the 51st Group. Chennault was then able to put 
fighters on a string of fields from Heng-yang to Kweilin in the Hsiang valley. 
Some missions were on occasion staged through newer fields 200 miles east of 
Kweilin. Sixteen P-51's, battered but still flyable, joined Chennault. Because 
his air-raid net and newly formed radio intelligence teams were working better 
than ever, Chennault found his fighter pilots spending more and more time 
over the Japanese bases in central China. Air superiority was changing hands, 
and perhaps, not too late in the day. 28 

Thanks to the arrival of B-25 pilots well schooled in skip-bombing tactics 
(first publicly demonstrated in the Battle of the Bismarck Sea, 2-6 March 
1943) and the wide dissemination of the technique among Chennault's crews, 
the 11th Bombardment Squadron (M) and the 2d Bombardment Squadron 
(Chinese- American Composite Wing) intensified the antishipping campaign. 
During October, the Fourteenth Air Force sank about 5,000 tons, and 17,372 

27 Stilwell's Mission to C7>/«dTc^EO_ 

28 ( 1) Stilwell's Mission to Unna f^mrrZ^ ( 2 ) Lt Gen Albert C. Wedemeyer's Data Book, Item 
8. OPD Copy, OCMH. (3) Ralph G. Hoxie, Frederick Ericson, and Robert T. Finney, History of 
the Fourteenth Air Force, MS. (Hereafter, Fourteenth AF History.) USAF Hist Div. 



tons in November. Simultaneously, the Japanese stirrings in northeast Burma 
that preceded the enemy drive on India gave the 308th Bombardment Group 
(H) a chance to drop bombs on Burmese targets while it engaged in its routine 
ferrying of supplies. 29 

The Fourteenth Air Force was now just able to reach beyond China's bor- 
ders for a few blows at Japan or Formosa. Heavy bombers and long-range 
fighters were on hand. Staging fields near the coast were ready, and one of 
them, Suichuan, was stocked to permit a mission or two to be flown from it 
against Kyushu or Formosa. On 31 October a photographic reconnaissance air- 
craft from Suichuan photographed a major Japanese shipping concentration in 
the Sasebo-Nagasaki area. Chennault at once asked Stilwell for permission to 
strike it, and Stilwell as promptly relayed the question to Gen. Henry H. 
Arnold, Commanding General, Army Air Forces. 30 Arnold congratulated Chen- 
nault on his aggressive spirit, but forbade attacks on Japan. Presumably, he did 
not desire to alert the Japanese home defenses by a few sporadic attacks when 
the B-29's were being prepared for a sustained effort." 1 

Though greatly disappointed because it was ordered to avoid Japan, and by 
implication also ordered to omit Phase III of the Chennault Plan, the Four- 
teenth Air Force immediately settled on Shinchiku, Formosa, site of a bomber- 
modification factory and a combat training center, as an alternate target. On 
Thanksgiving Day, in a meticulously planned, precisely executed tactical 
demonstration that lasted twelve minutes, 8 P-5lA's, 12 B-25's, and 8 P-38's 
scored 42 Japanese aircraft in one pass over the target. Exploiting the Japanese 
embarrassment, B-25's, working in pairs, employed their skip-bombing tactics 
to account for several cargo ships in the Formosa Strait during the next fort- 
night. 32 

By November the concentration of aircraft strength in east China brought 
into sharp focus the long-neglected line of communications from Kunming 
eastward to the Hsiang valley fields and beyond. November saw the Fourteenth 
Air Force receive tonnage at a level (4,700 tons) that Chennault had hoped to 
see achieved six months before. But tonnage at Kunming was just the begin- 
ning. The line of communications from Kunming to Chennault's forward 
fields, the eastern line of communications (known in CBI as the ELOC), was 

- 9 (1) Fourteenth AF History. (2) Wesley Frank Craven and James Lea Cate, eds., The Army 
Air Forces in World War II; IV, The Pacific: Guadalcanal to Saipan, August 1942 to July 1944 
(Chicago, 1950), Ch. 16, pp. 529-32. (3) Japanese Studies in World War II (hereafter, Japanese 
Stnd v V 116 The Imperial Japanese Navy in World War II, 1941-1945. OCMH. (See 

,u ( l ) Fourteenth AF History. (2) Rad M 18NE, Chennault to Stilwell and Arnold, 4 Nov 43. 
Item 1181, Bk 4, JWS Personal File. (3) CM-IN 1668, Stilwell to Arnold, 3 Nov 43. 

u (1) CM-OUT 1413, Arnold to Stilwell for Chennault, 4 Nov 43. (2) Compare the British 
revelation of the tank in World War I by its use in penny packets in the Somme fighting; the com- 
parable German blunder of introducing gas on a small scale. The combination of surprise and mass 
can be overwhelming, where the introduction of a new tactic or a new weapon in a small way 
merely warns the enemy to prepare countermeasures. 

" (1) Fourteenth AF History. (2) Japanese Study 116. 



a bottleneck. Currently operated by the Chinese— for it will be recalled the 
Generalissimo was Supreme Commander, China Theater, and U.S. resources 
in China were limited to what came over the Hump— the ELOC's estimated 
capacity of 1,500 tons a month barely supported two fighter squadrons in east 
China from May 1943 on. Now, with three more fighter and two medium 
bomber squadrons in east China, the ELOC had to increase its capacity, and 
Chennault commenced to stress its problems. 33 

Chennault's attack on Formosa had been made while several staff officers 
from China Expeditionary Army, Lt. Gen. Shunroku Hata commanding, had 
been visiting Imperial General Headquarters to prepare and co-ordinate future 
plans. With the campaign in the Solomons and New Guinea running against 
the Japanese, Imperial General Headquarters had rejected a plan to destroy the 
Generalissimo's government that General Hata had proposed. Instead it had 
told his representative about 17 October that beginning in December 1943 the 
Japanese 3d, 13th, 32d, 22d, 35th, and 36th Divisions would be sent from China 
to face the Americans in the Pacific, and that in the spring and summer of 1944 
five more of Hata's divisions would assemble in their present occupational areas 
to form Imperial General Headquarters' general reserve. Hata, as of 1 August 
1943, had 620,000 men, formed into 1 armored and 25 infantry divisions, plus 
12 brigades (of five infantry battalions each). Of the 25 infantry divisions, 

5 were Class A, with three battalions of artillery (36 cannon), 5 were Class B, 
with two battalions of artillery (24 cannon), and the remaining 15 divisions 
had no artillery. The projected transfers would take away 4 of Hata's Class A 
divisions, 1 Class B division, and 1 Class C division. This would cut deeply into 
Hata's strength, for the replacements would be raw troops that he would have 
to organize and train as units. So the mission Hata was now given probably 
reflected an appraisal of his current situation. He was ordered in October 1943 

. . to maintain security in the occupied areas. Enemy air forces were to be 
attacked at all times in order to prevent their making raids on the Japanese 

In November, Hata's liaison officers were told, by the Chief of the General 
Staff among others, that the Allied air force in China was disturbing Japanese 
sea communications and had to be neutralized or destroyed. Then came Chen- 
nault's Thanksgiving Day attack on Formosa. A staff officer of China Expedi- 
tionary Army, returning to his post in China 3 December 1943, told Hata that 
Imperial General Headquarters was "very disturbed" by the attack on Formosa 
and had begun to study the advisability of destroying the Allied airfields in east 
China and reopening the Canton-Hankow railway. 34 

Imperial General Headquarters soon decided to call a halt to the current Jap- 
anese operations in China, which had been launched early in October. A few 

33 (1) A full treatment of the ELOC is in Bykofsky MS. (2) Ltr, Chennault to Wedemeyer, 

6 Jul 45. WDCSA 091 China, 15 Aug 45. 

34 (1) Japanese Study 129, pp. 2, 11-13, 12, 17-18. (2) Ltr, Col Preston J. C. Murphy to 
Ward, 22 Oct 52. OCMH. 



weeks before the attack on Formosa the Japanese 11th Army had been moving 
across the Tung-ting Lake area toward Chang-te. Its ostensible mission was to 
disrupt Chinese troop concentrations in the VI and IX War Areas and to divert 
Chinese forces from Yunnan, but the primary objective was to seize rice. Under 
cover of bad weather, the 11th Army had completed its approach while the 
Fourteenth Air Force was forced to remain on the ground. When the weather 
cleared at the end of November, fighter squadrons at Heng-yang and Ling-ling 
had been on constant call to support the Chinese divisions defending Chang-te. 
On 4 December Chang-te fell, but apart from its value as a rice center the town 
held no further attraction for 11th Army. Once the VI and IX War Areas had 
been shaken up, the 11th Army withdrew. Imperial General Headquarters had a 
more important mission for it and wanted the 11th to take its place in the mass 
of maneuver Tokyo now decided to assemble. 

As the last shots were being fired around Chang-te, the highest officers in 
Imperial General Headquarters, the War Ministry, and the Naval General Staff 
were watching or actively participating in far-reaching map maneuvers, bear- 
ing the code name TORA. TORA was planned to test current hypotheses on 
the intentions and capabilities of Japan's enemies, to suggest a strategy with 
which to counter American plans in the Pacific, and to examine "the merits, 
the planning, and the military strength" involved in taking Kweilin and Liu- 
chow in east China. When TORA ended it was agreed to take the east China 
air bases and thereby "check" Allied air attacks from China. Plans and prep- 
arations were soon under way. 35 

The battles around Chang-te were most significant in the Fourteenth Air 
Force's analysis of its position in China. Operations over Burma in 1942 con- 
vinced the Fourteenth Air Force in May that air power had kept the Japanese 
from crossing the Salween. It further believed that its later sweeps over Burma 
had so disrupted their supply system that the Japanese were unable to prepare 
an offensive across the Salween. However, the Japanese had not contemplated 
crossing the Salween. In May 1943 the Fourteenth Air Force had supported 
the Chinese troops along the I-chang Gorge. The Japanese had fallen back 
from their most advanced positions, and Chennault had claimed a decisive vic- 
tory. Again, the Fourteenth Air Force's intelligence had misread the Japanese 
plans. Now, after Chang-te, the Fourteenth Air Force believed that it had 
created a successful technique for tactical air support of Gen. Hsueh Yueh's IX 
War Area troops. Close co-operation between the Fourteenth Air Force and 
General Hsueh was important because his war area guarded the northern ap- 

(1) Statement of Col Takushiro Hattori, 21 Jul 49, in Statements of Japanese Officials in 
World War II, Vol. I, pp. 370-72. OCMH. (2) Japanese Studies 77, 78, 129- (3) Imperial 

General Headquarters Army Order 921, 24 Jan 44 (hereafter, IGH Army Order ), GHQ, 

Far East Comd, Mil History Div, Imperial General Headquarters Army Orders, Vol. II. OCMH. 
(See Bibliographical Note.) (4) Statement of Marshal Shunroku Hata, Aug 52. in Japanese 
Officers' Comments (hereafter, Japanese Officers' Comments), Incl 2. OCMH. (See Bibliographical 
Note.) ' 



proaches to Chennault's fields. The seeming success of combined Sino- Amer- 
ican operations before Chang-te moved the Chinese to permit the Fourteenth 
Air Force to expand its radio and intelligence net within the IX War Area. 36 

Chennault's Proposals for 1944 

Chennault had been present in October 1943 when Mountbatten, Stilwell, 
and the Generalissimo discussed the Quadrant decisions on breaking the 
blockade of China. It appeared to him that "no definite decisions had been 
reached with respect to the Burma campaign." Such a campaign would, he 
thought, make it very difficult to launch his 1944 China offensive and so he 
again hastened to present his case to his Army superiors before, in his opinion, 
the United Nations were firmly committed to the Burma campaign. Chen- 
nault also appealed to the President, telling him that it had been impossible 
for the Fourteenth Air Force to launch an effective air offensive in 1943. Only 
now were his reinforcements coming forward. 3 ^ 

The Chennault proposals of October 1943 were basically similar to those 
of the previous April, but there was one interesting addition. 38 If the General- 
issimo was prepared to contemplate an offensive by the Chinese Army against 
Hankow or Canton, the necessary tactical air support operations could be 
meshed into the over-all plan. Where the April proposals had of necessity been 
hastily drafted by Chennault en route to the Trident Conference, those of 
October showed a good deal of careful preliminary staff work. It was most 
notable that the line of communications to the Fourteenth Air Force's fields 
had been carefully studied, and that detailed suggestions for its improvement 
were included. In April, this aspect had been omitted. 

For operational purposes, Chennault divided China into eastern and west- 
ern areas, on the line of the 108th meridian. Weather phases in the two areas 
are markedly different and influence the timing of operations. From January 
to June, weather is good in the west, then grows steadily worse in summer. In 
the east, flying weather is good from July to December. Therefore, from Jan- 
uary to June, Chennault proposed to fly most of his missions in west China in 
support of the Burma campaign and in defense of the Hump airfields. From 
July to December he would concentrate on the goals he had so often set forth, 

36 ( 1) See Stilwell' s Mission to China, Ch j VIII j m jlXl (2) Se < \ Ch. X| below. (3) Claire L. 
Chennault, Way of a Fighter (New York, 1949), pp. 256-64. *4) Japanese Study 77. (5) 
Chennault thought the Japanese retreat proved his contention that "existing Chinese forces with 
adequate air support can stop any Japanese advance which is not so great as to weaken their 
defense elsewhere." Ltr, Chennault to Hopkins, 27 Dec 43. Bk VII, Hopkins Papers. 

37 (1) For a discus sion nf Chennault's corresponding directly with the President, see St H well's 
Mission to China, Chs. |IX| an J X| (2) Ltr, Chennault to Roosevelt, 28 Oct 43. Bk VII, Hopkins 
Papers. Two days later™tne CnTnese Army in India engaged the Japanese 18th Division, opening 
the North Burma Campaign. (3) Memo, Wedemeyer for Mountbatten, 10 Nov 43, sub: Chennault 
Plan for Air Offensive From China. Attached is Chennault's plan, dated 17 October 1943. Folder, 
Chennault Air Plan, CT 39, Dr 1, KCRC. , , 

38 Chennault's April 1943 proposals are in Stilwell 's Mission to China, |Ch, IX. | 



destruction of Japanese air power and merchant shipping. Perhaps, in these 
provisions, Chennault was offering a compromise to Stilwell, in line with the 
promise he made in October to support StilwelPs plans. 
To supply his 1944 operations, Chennault asked for: 

Eastern Area Western Area 
(Tons) (Tons) 

January 1900 6900 

February 2100 6900 

March 2300 6900 

April 2700 5400 

May 4500 3400 

June 6200 2700 

In addition, Chennault wanted a reserve of 20,000 tons accumulated in east 
China before his operations opened up in full scale, which was to be in July. 
To acquire these supplies, Hump tonnage would have to average 12,000 tons in 
the first six months of 1944, and the Fourteenth Air Force receive every pound 
of it. The plan stressed the importance of insuring delivery of supplies to the 
east China bases and gave a carefully worked out plan to insure this. Improve- 
ment of the eastern line of communications was estimated to require an initial 
4,000 tons over the Hump in January 1944 plus 500 tons a month for the rest 
of the year. For aircraft, Chennault wanted by fall of 1944 to have 6 fighter 
groups, 2 medium bomber groups, and 3 heavy bomber groups. 39 

Chennault dismissed the possibility of a Japanese ground attack on his air 

6. Previous proposals for a China Air Offensive have been objected to, on the grounds that 
"in china the Japanese can go anywhere and take any objective they want," and can there- 
fore capture and destroy our airbases. This theory is not founded on fact, for the following 

A. The proponents of the theory think exclusively in terms of a time when the Japanese 
domination of the air was rarely challenged by a single Chinese aircraft. In the event that 
plans for a China Air Offensive are approved, the Allied armies will have air supremacy on 
their side. Chinese troops have shown that with the encouragement of even a modicum of 
air support, they are far better able to withstand the Japanese. It may be presumed that Jap- 
anese troops, suddenly confronted with the transfer of air supremacy to their enemies, will 
also fight less well. 

B. The proponents of the theory also forget that in the whole course of the war in china, 
the Japanese Armies have never successfully penetrated more than 100 miles beyond their 
major supply lines. In the CHEKIANG campaign in the winter [sk] of 1942, air bases were in 
fact captured. But the campaign was long, employed very considerable forces, and was fought 
by the Japanese with the double advantage of complete domination of the air and proximity 
to their supplies. Weakness of enemy supply enabled unaided Chinese troops, fighting with- 
out air support, to frustrate no less than five attempts to capture the line of railroad which 
formerly connected canton and hankow. Two of these attempts were made from the 
South and three from the North; all were in substantial and at least two were in great force. 

(1) Stilwell' s Mission to Cbtna \Ch. X*| (2) Plan cited |n. 37 (3)"] 



C. Establishment of air supremacy by a China Air Offensive will further increase the Jap- 
anese weakness of supply. More than 95 per cent of the supplies of the Japanese Armies in 
China from the yangtsze river region Southward are carried inland on small, slow com- 
monly unarmed river craft. As an experiment, traffic on the yangtsze river has been al- 
most totally interrupted for a period of a week with a force numbering not more than two 
score aircraft. With sufficient forces, adequately supplied, this interruption can be consist- 
ently maintained, on other waterways as easily as on the yangtsze. The enemy has no 
practicable substitute supply lines for operations on any scale. 

Thus is it seen that a major enemy effort, both on the ground and in the air, would be 
necessary to attempt to interrupt a China based Air Offensive. Ships carrying personnel and 
equipment for such an effort into china would be exposed to our land based bombardment. 
And to make such an effort, the Japanese would be forced to abandon essential commitments 
elsewhere. 40 

There were several mentions in the plan of possible ground operations by 
the Chinese to reoccupy portions of China. It was suggested that in the last six 
months of 1944 support might be given to Chinese troops in operations against 
the Canton-Hong Kong and I-chang-Hankow-Nanchang areas. After remark- 
ing tersely that "adequate Chinese ground forces must of course be maintained 
in the airbase areas" the plan went on to state that in the last half of 1944 
"enemy garrisons in china . . . will be gravely weakened" by air attacks on 
Japanese supply lines in China. If the Generalissimo thought Chinese ground 
forces "available" the Japanese garrisons in China might be attacked. If any 
such operations were to be undertaken, then the Chinese should be given "cer- 
tain key items of equipment which the Chinese Ground Forces now lack." 41 

No document has been found to give Stilwell's reactions to Chennault's 
October 1943 proposals. However, at the time he summarized his impressions 
of the "net worth" of Chennault's 1943 efforts as "a few Jap planes knocked 
down [but] Japs are still in China." 42 

Appraising the plan for Admiral Mountbatten, Maj. Gen. Albert C. Wede- 
meyer, as SEAC's deputy chief of staff, commented that it was "essentially the 
same as one previously submitted by General Chennault several months ago." 
Wedemeyer noted that the Chennault October 1943 proposals did not allot 
any Hump tonnage for equipment of the Chinese armies that were to protect 
the east China air bases. Wedemeyer believed that Chennault greatly under- 
estimated the tonnage he would need for his operations and that his actual re- 
quirements were in excess of the "Air Transport Command capacity with 
resources now in sight." In his conclusions, Wedemeyer advised Mountbatten 
that a "practicable" increase of the Fourteenth Air Force was highly desirable 
and should be supported after it had been weighed in connection with other 
demands. He warned that "the ASSAM line of communications and the Air 
Transport Command cannot support the air force visualized in the Chennault 

40 Plan cited ln. 37(3)1 

41 Ibid. 

42 Stilwell's Mission to China^^^>%^ 



plan and concurrently supplies for the rehabilitation of Chinese internal lines 
of communication and for the Chinese ground forces." In Wedemeyer's opin- 
ion, the fastest way to begin intensive air operations in China was to open a 
"land and/or sea route" to the Chinese airfields. 43 

What must be taken as the CBI Theater decision on Chennault's October 
1943 proposals was the response of General Stratemeyer's headquarters, Army 
Air Forces, India-Burma Sector, on 16 November 1943. Though he liked the 
careful phasing of Chennault's plan, Stratemeyer in effect rejected it with the 
remark that "we are not yet in a position logistically to support this plan at this 
time [Stratemeyer's italics} from a stores or a POL [petrol, oil, lubricants] 
standpoint." Stratemeyer raised two points which he believed would determine 
the scope of Chennault's operations in 1944. Aware that Chennault's Hump 
priority expired 31 October, Stratemeyer told the Fourteenth Air Force com- 
mander that he could not have the Hump tonnages he wanted, and that Stil- 
well was contemplating a division of Hump tonnage that would provide more 
for the Chinese Army. Stratemeyer's second point reflected the implications of 
bringing the B-29's to CBI, and practically assigned second priority in air op- 
erations to the Fourteenth Air Force. 

The President has indicated that the Twilight plan will be carried out and has cabled 
Chiang Kai-shek to give it his fullest support. It is believed, therefore, that the greatest good 
for the War effort could be obtained if our entire resources were thrown in gear to supple- 
ment the Twilight plan (including increased weather service in China) and, in addition, 
keep your operations rate in China against enemy shipping, defense of the air terminals and 
support of the Chinese ground forces at the highest rate possible under these logistical 
restrictions. 44 

Stilwell in September had wanted to readjust Hump priorities to give Chen- 
nault a flat 40 percent of Hump deliveries but Marshall had warned him to be- 
ware of the political implications of such an act. 45 Now, a few weeks later, 
Stratemeyer signaled the end of Chennault's priorities by merely stating that 
Stilwell "desires" to readjust Hump priorities without explaining whether this 
was an order of the theater commander or had War Department sanction. 46 
The records available reveal no protest from Chennault or attempt to have 
Stratemeyer overruled by higher authority. There had been significant changes 
in the identity and status of those who supported Chennault's views, and the 
airman may have considered that without support from T. V. Soong and 
Madame Chiang Kai-shek he could not hope to repeat the coup of the 
Trident Conference. 47 

43 Memo cite d n. 37 f 3^.1 

44 Memo, Stratemeyer for Chenna ult. 16 Nov 43, sub: Operational Plan for Air Offensive in 
China, 1 Jan-31 Dec 44. Ltr cite d n. 33(2YI App. II, Item 21. 

45 (1) CM-IN 17227, Stilwell to Marshall, 24 Sep 43. (2) CM-OUT 12435, Marshall to 
Stilwell, 27 Sep 43. 

46 Memo cited n. 44. ■ ■ 

47 Stilwell 's Mission to China\ Ch. X. | 



Building an East China Army 

No less than Chennault, Stilwell had a program for east China, and their 
competing claims on Hump tonnage did much to explain the acrimony be- 
tween the two men. Like Chennault's, Stilwell's program was begun in 1942, to 
carry out his War Department orders to improve the combat efficiency of the 
Chinese Army. As part of his over- all program to rebuild the Chinese Army, 
Stilwell in early 1943 suggested to the Generalissimo that thirty reorganized 
and re-equipped Chinese divisions should be placed in east China. Stilwell re- 
ceived War Department support for the project, and lend-lease for the thirty 
divisions in east China (the "Second Thirty" they were sometimes called to 
distinguish them from the "First Thirty" of the Thirty Division Program or 
the Y-Force in Yunnan plus the Chinese Army in India). But the Chinese 
waited many months, in fact until after the Quadrant Conference (August 
1943), before they would consider the project. 48 

On 1 September, General Ho, the Chinese Army's Chief of Staff, startled 
Stilwell by proposing a tentative list of forty-five divisions, thinly spread from 
the Yellow River to the Indochinese border, from which a "C-Force" (later 
called Zebra Force) could be equipped, trained, and then shaped into a useful 
force. Stilwell was jubilant, and wrote: "victory! That commits them to the 
training scheme. Subject to change, of course, but what a struggle that has 
been. . . . If the Japs let us alone, we may put it over!" Stilwell saw the mis- 
sions of Zebra Force as: (1) to defend the east China airfields; (2) to drive the 
Japanese from the Yangtze valley; (3) to co-operate in any Allied plans to land 
on the China coast. 49 

Attempts to create the Zebra Force quickly revealed basic differences be- 
tween the Chinese and American approaches to the problem of creating a force 
in east China. The Americans wanted to bring the chosen divisions up to 
strength, weed out incompetent officers, train the soldiers, and then issue 
shining new lend-lease equipment. The Chinese wanted to receive the equip- 
ment, then talk about training and reorganization. To Stilwell, this attitude 
toward Zebra Force was exactly like that the Chinese had taken toward 
the Yunnan divisions or Y-Force. Nevertheless, after the Generalissimo ap- 
peared more interested in Army reform and more co-operative in October 1943, 
Stilwell was able to send Brig. Gen. Thomas S. Arms, of the Y-Force Infantry 
Training Center, to establish a similar training center at Kweilin in east China. 
To assist in training and liaison, 2,213 officers and enlisted men were sent to 
CBI. 50 

48 Stilwell's Mission to China, Chs. | VIl| an J X.| 

4 9 (1) Z-Force Order of Battle, jTTXug 4%~AG (Z-FOS) 381, KCRC. (2) Theodore H. 
White, ed., The Stilwell Papers (New York: William Sloane Associates, Inc., 1948), p. 219. (3) 
CM-IN 6110, Stilwell to Marshall, 2 Sep 43. (4) CM-IN 2881, Stilwell to Marshall, 5 Nov 43. 

50 (1) History of Z-FOS, 1 Jan-31 Oct 1944. OCMH. (2) Ltr, Stilwell to Arms, 8 Nov 43, 
sub: Activation of ITC at Kweilin. AG (Z-FOS) 320.2, KCRC. (3) History of India-Burma 
Theater, 1944-1945, I, 55. OCMH. 




At the time, Stilwell asked the War Department to review its 8 July 1943 
program for Chinese lend-lease (as distinct from the 1944 supply program). 
The July program assigned lend-lease to the First Thirty, but only 10 percent 
of the equipment needed by the Zebra Force, which token shipments were to 
be used for instructional purposes. Stilwell wanted a firm War Department 
commitment to arm the Zebra Force in full. Faced with this request, the War 
Department studied it in the light of the Quadrant decisions, the limitations 
on the capacity of the Assam line of communications and the Hump airlift, and 
the emerging strategic concepts for the conduct of the war against Japan, which 
were beginning to set a lower value on China's co-operation. After studying 
the problem, the War Department decided that it would review its policy in 
the light of the degree of co-operation Stilwell might in the future receive from 
the Chinese, and reaffirmed its policy of confining its logistical support of 
Zebra Force to the 10 percent figure. 51 

In November and December 1943 the Chinese gave evidence of being 
genuinely interested in the Zebra Force project. The Kweilin Infantry Train- 
ing Center was opened, with the Generalissimo and Gen. Pai Chung-hsi of the 
National Military Council as its honorary commanders. The instruction, 

51 Memo for record, 30 Oct 43; Memo, Hull for ACofS, G-4, WDGS, 15 Nov 43, sub: 
Establishment of Policy of Supply of Combat Type of Lend-Lease Materials to China. OPD 400 
CTO, A47-30. 



which followed a course approved by the Chinese, was carried on by 107 U.S. 
officers and 71 enlisted men. Several buildings were turned over to the training 
center by the Chinese, and the Americans contracted for many more. 

Though infantry and artillery training was stressed, engineering, veterinary, 
medical, and signal courses were offered. Training equipment, initially flown 
from Kunming, was furnished by the U.S. Army. The infantry course lasted six 
weeks, of which the first three stressed weapons instruction and target practice. 
The last half of the course was devoted to tactics, taking the individual officer 
progressively to the regimental level. Eighty-nine interpreters gave invaluable 
aid. The student officers were placed in training regiments, each with twelve 
companies of 100 students per company. 

After the first infantry class graduated on 15 December 1943, General Arms 
estimated that by 1 May 1944, 4,800 infantry officers and 2,730 officers and men 
from other arms and services would have graduated. Manifestly, this figure 
would depend upon the continued willingness of the Chinese to send students 
to the training center, a willingness that might not survive the War Depart- 
ment decision to keep equipment for Zebra Force at the 10 percent level. 52 

The Allied Command Structure in North Burma 

Improvements in the Assam line of communications, the decision to place 
a force of B-29's in CBI, and attempts to strengthen the Allied position in east 
China, were, as it developed, both background and accompaniment to the pre- 
mature and unscheduled opening of combat in north Burma. Certainly, the line 
of communications had to be renovated before the Allies could be sure of 
themselves, while the presence of the B-29's affected the course of things in 
China, but the principal event in fall 1943 was the opening of the North Burma 
Campaign. What had been contemplated were roughly simultaneous attacks 
from India and China into Burma but this was not to be. 

The Allied forces in north Burma operated under a very complicated system 
of command. North Burma was within SEAC's boundaries, and therefore the 
chain of command began with the Supreme Allied Commander, Admiral 
Mountbatten. As acting Deputy Supreme Allied Commander, Stilwell was sec- 
ond man in SEAC, but as Commanding General, Chinese Army in India, he 
commanded a pair of Chinese divisions which had to be fitted into the organ- 
ization. Normally, as a corps commander Stilwell would have been in the chain 
of command which ran from Admiral Mountbatten to Gen. Sir George Giffard 
(General Officer Commanding-in-Chief, Eleventh Army Group) to General 
Slim (General Officer Commanding, Fourteenth Army) to Stilwell. (See Chart 
1 1. jj Though Stilwell's initial reaction to Mountbatten was very favorable, he had 
no confidence in General Giffard and refused to occupy any place in the com- 
mand structure that might permit Giffard to exercise control over his opera- 

52 (1) History cited |n. 50(TT1 (2) Z-Force Journal. KCRC. 



tions. Instead, Stilwell offered to serve as a corps commander under General 
Slim, for whom he had the greatest faith and respect. 

Stilwell's service under Slim was to continue until Stilwell's forces reached 
the Kamaing area, at which point he was to regain independence of action. It 
was assumed that when Stilwell's Chinese from Ledo reached Kamaing they 
would be near a juncture with Chinese troops from Yunnan Province. On join- 
ing, the two Chinese forces would be under Stilwell, who in turn would be 
directly under Mountbatten with no intervening echelons of command. 

Informed a few weeks later of Stilwell's agreeing to forego rank and serve 
under a junior he admired, General Marshall offered an interesting appraisal of 
Stilwell on the eve of the latter's greatest triumphs as a field commander. Mar- 
shall wrote Mountbatten: 

You will find, if you get below the surface, that he wants merely to get things done with- 
out delays and will ignore considerations of his own personal prestige or position so long as 
drive and imagination are being given to plans, preparations and operations. 

Frankly, I have found him uniformly through long years of personal command relations, 
irritating and intolerant of slow motion, excessive caution, and cut-and-dried procedure. On 
the other hand, he will provide tremendous energy, courage and unlimited ingenuity and 
imagination to any aggressive proposals or operations. His mind is far more alert than al- 
most any of our generals and his training and understanding are on an unusually high level. 
Impatience with conservatism and slow motion is his weakness — but a damned good one in 
this emergency. 53 

Stilwell's command relation to the Generalissimo in north Burma was ob- 
scure. In 1942 the Generalissimo's attitude had suggested that Stilwell would 
command the Chinese Army in India only until operations began. By Septem- 
ber 1943 the most probable Chinese commander was Lt. Gen. Chen Tung-kuo, 
commanding the New First Army headquarters, of whom a memorandum in 
Stilwell's personal file stated: "This officer may be capable, but he has not yet 
demonstrated the fact ... no concern about the basic needs of training, which 
he does not understand. He is not interested to learn from the bottom up. . . . 
As a matter of fact, there is no need whatever for any army staff, as long as the 
Chih Hui Pu exists. . . ." 54 The Chinese 38th, 22d, and 30th Divisions were 
under the New First Army, but since that headquarters was regarded as super- 
fluous by t he Americans, Stilwell sought to have it removed from the scene. 
(Chart 3) | 

The headquarters of the Chinese Army in India was potentially Stilwell's 
field headquarters, though Brig. Gen. Haydon L. Boatner, chief of staff and 
deputy commander of the Chinese Army in India, was in charge. The staff was 

53 (1) Opn Instr 1, Eleventh Army Group to Comdr, Fourteenth Army, sub: Opns in Burma, 
1943-44. SEAC Info Bk, OCMH. (2) Rad CHC 1111, Stilwell to Maj Gen Daniel I. Sultan, 27 
May 44; Rad RE 89, Sultan to Stilwell, 15 Jan 44. Items 188, 17, Bk 6A, JWS Personal File. 
(3) Ltr, Mountbatten to Marshall, 16 Jan 44; Quotation in Ltr, Marshall to Mountbatten, 26 Jan 
44. Case 297, OPD 381 Security, A47-30. 

54 Memo [Brie Gen Havdon L. Boatner?], 24 Sep 43, sub: Personnel. Stilwell Numbered File 
[SNF] 16. (See |Bibhographical Note!) 

Chart 1—Cmn Hui Pu 


11& tPO. <» 

23 JANUAftTT u« 



9*33*1 », ™ rti-. 

u UTrtOBiiu ml nor icmnTED 
% MM Kiwn Tt, It*t»«Tn. 

* njnns " <jm> 

* mm nee » ftwmt 



L "iff. I 

!TTH V j H »T KH 

'itTV- ■ — Q«o" »aSii. ! 

2D- 1 -HEZTtmek? j25>j>j«^i] 


H!Lf> 31 . — i 


rrii*T^n»"y" ^3 n*r ~i 
r»"TWTt*LD'}wi '** ~i 


Elf VH* y£s°!I™k7Z. Z3 

■ »^ _j ■ - , - 

■I - "L ' . 

j i m7"?q i «: (. f» " ' I7i I 

rja' ^^ i »?-tr* i "' '1771 
riSiiL t * * "777! 

U.»_J !*-.»} jC" 

lm-Zj E?D S?I3 



American, but the headquarters was regarded legally and diplomatically as 
Chinese. All orders to Chinese units initially went on Chih Hui Pu letterheads, 
validated by StilwelPs chop at the bottom as commanding general. When the 
38th Division moved up to the Ledo area in April 1943, Chih Hui Pu split into 
a forward and rear echelon. The Forward Echelon was under General Boatner, 
who relieved General Wheeler and the Services of Supply (SOS) of respon- 
sibility for the forward area. Brig. Gen. William E. Bergin took charge of the 
Rear Echelon. 55 

Before General Boatner could play his part in the forthcoming campaign, 
he had to survive a Chinese attempt in September to remove him, and a sug- 
gestion that Lt. Gen. Sun Li-jen, commanding the Chinese 38th Division, have 
a free hand in the approaching operations. Whatever the abstract merit of hav- 
ing General Sun— one of the few Allied commanders to emerge from the First 
Burma Campaign with enhanced reputation — practically in command of the 
North Burma Campaign, General Sun's hint that he should take over was cast 
in a form that made it unacceptable. Sun's letter to Stilwell took so dismal a 
view of the campaign, referred with such gloomy relish to the Japanese strength 
and the difficulties of the terrain, as to suggest he approached the fight with 
extreme reluctance. Stilwell could hardly have entrusted the campaign to a man 
who had gone on record as having very little faith in it. But Sun had the back- 
ing of the Generalissimo, at least for Sun's wish that Boatner go, and the Gen- 
eralissimo ordered Stilwell to relieve Boatner. Stilwell replied that the com- 
mander's deputy, Boatner, must be able to command both the Chinese and the 
U.S. service troops who would be supporting the drive. Stilwell knew of no 
Chinese officer able to fill that role. If the Generalissimo did, let the officer be 
named and a simple, practical test be given to determine the question. The 
issue was dropped and Boatner kept his post. 56 

The American personnel of Chih Hui Pu were concurrently 5303d Head- 
quarters and Headquarters Company (Provisional) Combat Troops. They func- 
tioned as an American headquarters for the American service and medical units 
in the Chinese Army, which had few such units. 57 As American, British, and 
Indian combat and service units entered north Burma, they were attached or 
assigned to the 5303d and its successor headquarters, with the exception of Stil- 
well's regiment of U.S. jungle-trained infantry (code name Galahad) which 
on arrival was attached to the Chinese Army in India and later to the American 
headquarters. 58 There was no boundary between the communications and the 

" (1) Ramgarh: Now It Can Be Told (Ranchi, India, 1945), p. 14. (2) Ltr, Stilwell to Boatner 
and Whe eler. 6 Apr 43. Item 214. Bk 3. IWS Personal File. 

56 ( 1) | Stilw ell' s Mission to Chi na, Ch. XJ (2) Ltr, Sun to Stilwell, 17 Aug 43, quoted in Dr. 
Ho Yung-chi, The Big Circle (New York: The Exposition Press, 1948), p. 64. (3) Memo, Stilwell 
for Generalissimo, 27 Sep 43, Item 233, Bk 3, JWS Personal File. (4) Rad BURSAM OT-11, 
Stilwell to Boatner, 21 Sep 43. Item 1027, Bk 4, JWS Personal File. Stilwell suggests that the 
whole thing is an attempt to remove him from the campaign via Boatner. 

57 Ramgarh: Now It Can Be Told. 

58 (1) GOs 11, 12, 14, Hq Rr Ech USAF CBI, 29 Jan, 5 Feb 44. (2) GO 74, Hq USAF CBI, 
17 Jul 44. 



combat zones, with the result that SOS in the forward zone was duplicating the 
work of the combat troops in supporting the Chinese. 59 An important member 
of Chih Hui Pu was Capt. C. E. Darlington, who before the war had been dis- 
trict commissioner of the Hukawng Valley for the Government of Burma. Dar- 
lington was completely familiar with that section of north Burma, had the 
loyalty and respect of the Kachin tribesmen who inhabited it, and was an un- 
failing source of advice and information. 60 

The Chinese Forces 

The Chinese 38th and 22d Divisions were commanded by General Sun and 
Lt. Gen. Liao Yao-hsiang respectively. General Sun, a graduate of the Virginia 
Military Institute, was appraised as: "A good field soldier, courageous and de- 
termined. Argumentative, insistent on doing things his way; but much con- 
cerned over the well-being of his unit. A capable commander." General Liao, 
a graduate of St. Cyr in France, was judged: "A good field soldier, courageous 
and determined. He has faithfully complied with all training directives, and as 
a result his division is now fully as good as the 38th, which had a long start 
ahead of him. A capable commander." 61 

When the 22d and 38th completed their training at Ramgarh, the 38th 
numbered 946 officers and 11,388 enlisted men, and the 22d, 857 officers and 

10,439 men. (See Chart 3.) With service and army troops, the Chinese Army 
in India totaled 2,626 officers and 29,667 enlisted men. Replacements were pro- 
vided and the two divisions were actually slightly larger at the end of 1943- 
The army and service troops included an "excellent" antiaircraft battalion, a 
motor regiment which worked in the Ledo area, an engineer regiment which 
built most of the Shingbwiyang airstrip, an animal transport regiment, and two 
porter units to carry stretcher cases. 62 

The work horse of the campaign was the infantry regiment, with 129 of- 
ficers and 2,642 enlisted men at Table of Organization (T/O) strength. It had 
three battalions, each with three rifle companies and one machine gun com- 
pany. The battalion had 27 Bren guns, 51 submachine guns, and 18 60-mm. 
mortars. 6 ^ The machine gun company had 8 heavy .30-caliber machine guns. In 
the regiment for fire support were a mortar company with 12 81-mm. mortars 
and an antitank company with 8 37-mm. antitank pieces. There were two trans- 

59 Ltr, Hq 5303d (Prov) Combat Trs to CG, Fwd Ech USAF CBI, 2 Dec 43, sub: Responsibility 
of Comd and Opns Within Combat Zone. AG (NCAC) 323.3, NCAC Files, KCRC. 

60 Wkly G-2 Rpt, App., Hq 5303d Combat Trs; Ltr, Maj J. W. Leedham, Hq Combat Trs. 
NCAC Files, KCS£ 

61 Memo citec l n. 54. 1 

62 (1) Strength Rpt, G-l, Per Rpt, Hq Chih Hui Pu, 20 Jan 44. NCAC Files, KCRC. (2) Opn 
Rpt, 22d Div (Chinese), 8 Jan 44. Albacore Hist File, KCRC. 

63 Thanks to the initiative of Dr. Lauchlin Currie, administrative assistant to President Roosevelt, 
and to the generosity of Canada, the Chinese received from Canadian Mutual Aid programs Bren 
guns, Boreantjtank rifles, Bren gun carriers, and other British-type ordnance. See Stilwell's Mission 
to ChinalCh. I.l 



portation companies (one a pack transport), a signal platoon, a special service 
platoon, and medical and veterinary personnel. There were fourteen radios and 
plenty of wire equipment. 64 Division artillery was on the modest side because 
of terrain conditions and was attached to regiments as the situation required. 
As the campaign progressed, the demand for artillery support grew. As a result, 
by June 1944, one battery of 155-mm. howitzers and one of 105-mm. howitzers 
were firing in general support. Support by medium and fighter bombers was 
available on an increasing scale. Allied air superiority in north Burma was 

Because of the extremely difficult nature of the terrain in north Burma, ex- 
tensive reliance was necessarily placed on animal transport, such as mules, 
horses, and Indian ponies. Each division had about 1,000 animals. Forage for 
the animals was regularly supplied by air. 65 

A solution to the replacement problem had been arranged in that the 
Chinese Government had promised to supply replacements; the success of Stil- 
well in holding the Chinese to the promise would have a great deal to do with 
the progress of the campaign. Stilwell's staff believed that much of what they 
took to be the undue caution of Chinese commanders in the conduct of oper- 
ations arose from the lack of a functioning replacement system. The Americans 
believed that the Chinese commander whose force took casualties in battle suf- 
fered a proportionate loss of power and influence. Consequently, the Chinese 
appeared reluctant to embark on combat operations. A steady flow of replace- 
ments from China would do a great deal to ensure aggressiveness by Sun and 
Liao. 66 

To exercise command, Stilwell drew on his 1942 experiences and organized 
an American liaison net down to and including the Chinese battalions. Each 
Chinese division had a small staff of U.S. advisers, including supply, signal, 
medical, motor, and veterinary officers, under a colonel, who kept in touch with 
Chih Hui Pu through a division radio platoon and three radio teams. Chinese 
regiments had a liaison officer of field grade, with radio team; each battalion 
had a major or senior captain with radio team. 

It speaks well for both Americans and Chinese that over a period of months 
a modus vivendi was established, for liaison personnel were in a delicate situ- 
ation. They had no powers of command, and their attempts at persuasion and 
advice often jarred Chinese notions of face. Further, they were working with 
good Chinese troops under commanders who had had experience of fighting 
in Burma. It often must have been hard for young Americans on their first cam- 
paign to offer suggestions to seasoned veterans. Interpretation of Chih Hui Pu's 

64 T/O&E, Chinese Inf Regt, 16 Mar 43. AG (NCAC) 320.3, KCRC. 

65 (1) G-3 Rpt, Chih Hui Pu, 27 Jun 44. KCRC. (2) Capt. Edward Fisher, History of the 
Northern Combat Area Command (NCAC), CBI Theater and IB Theater, MS (hereafter NCAC 
History), App. 6, History of Air Dropping. OCMH. 

66 ( 1) Ltr, Col Hill to Ward, 2 Sep 52. (2) For examples of Stilwell's efforts to obtain replace- 
ments, see Items 1601, 1619, Bk 5, JWS Personal File. 



orders was a source of potential difficulty, not only interpretation in the nar- 
row sense but in that of conveying the spirit as well. Important clauses, in 
translation from English to Chinese, were always reinforced with parallel con- 
structions, so that one clause at least might be understood. Though no worse 
than what other Americans faced in the Southwest Pacific and a little better 
than what the Chinese bore with cheerful fortitude, conditions in the field for 
the liaison personnel were still hard and were a strain on those Americans who 
bore them. There was lacking the sense of fraternity with and close support by 
one's own people, supplies sometimes vanished, and hostile Chinese officers 
could make life most unpleasant. 67 

The American Force 

For the projected Burma operations, the War Department had at last sup- 
plied some U.S. infantry troops, though originally they were not intended to 
operate under Stilwell's command, and in number were far from the corps of 
which Stilwell had always dreamed. On 1 September 1943 General Marshall 
had directed shipment of about 3,000 volunteers to Asia to form three long- 
range penetration groups on the model of those commanded by Brigadier Orde 
Charles Wingate. The project was given the code name Galahad. Organiza- 
tion of Galahad began on 5 September and was complete on 20 September. 
Given the designation of 5307th Composite Unit (Provisional), the unit was 
formed of volunteers from the continental United States, the 33d Infantry 
Regiment on Trinidad, British West Indies, and from the Southwest Pacific. 

The War Department did not think that Galahad could be restored after 
action by receiving replacements. The War Department told theater head- 
quarters its conception was that the unit was provided for one major mission 
of three months' duration, whose close might find the unit so exhausted and 
depleted that its survivors would require three months' hospitalization and rest. 
Through rumor, this idea was conveyed to Galahad personnel in the some- 
what different form that after three months' combat duty they would be 
relieved. 68 

After disembarking at Bombay between 29 and 31 October 1943, Galahad 
passed under SEAC's operational control, and SEAC began to train the men 
according to Wingate's doctrines. This arrangement caused some protest from 
CBI Theater staff officers, who thought it might be taken as an admission that 

67 (1) History of CBI, Sec. II, Ch. I, pp. 8-1 1. (2) Ltr, Lt Col Trevor N. Dupuy to Ward, 12 
Sep 52. OCMH. (3) Records of incidents involving Chinese and Americans are to be found in 
NCAC Files, KCRC. 

68 (1) Memo, Marshall for Dill, 26 Sep 43. WDCSA (China), A45-466. (2) Rad WAR 3495, 
Marshall to Stilwell, 2 Oct 43. Item 1056, Bk 4, JWS Personal File. (3) Ltr, Office IG CBI to CG 
USAF CBI, 12 Jul 44, sub: Investigation Re 5307th Prov Unit. AG 333, NCAC Files, KCRC. (4) 
Min, JCS 107th Mtg, 18 Aug 43, Item 2. (5) Ltr, with Incl, Col C ;arles N. Hunter to Ward, 14 
Aug 51. OCMH. Colonel Hunter's personal correspondence file was attached to his letter. (6) 
Notes by Hunter on draft MS. (Hereafter, Hunter Notes.) OCMH. (7) Ltr, Lt Col George A. 
McGee, Jr., to Ward, 1 Sep 51. OCMH. (8) Notes by McGee on draft MS. OCMH. 



Americans did not know jungle fighting. But General Marshall's answer to this 
and other arguments of like tenor was the simple statement that all hands 
would have to eat some crow if Japan was to be beaten. 69 It may be surmised 
that the Chief of Staff believed no coalition could endure if one partner always 

That Galahad was under SEAC's operational control, rather than under 
CBI Theater directly, caused administrative difficulties, for there was no prior 
decision as to which headquarters would exercise administrative responsibil- 
ities. Nor was Galahad even activated as a unit until several months had 
passed. Initially, OPD had ordered Lt. Col. Charles N. Hunter, as "Command- 
ing Officer all Casual Detachments, shipment 1688," to prepare Galahad's 
personnel for their future duties. 70 On 13 November CBI Theater headquarters 
ordered Col. Francis G. Brink to be "officer-in-charge of training, Galahad 
project." Brink was ordered to deal directly with Rear Echelon headquarters at 
New Delhi on "all matters pertaining to administration and supply." 71 But 
this letter did not activate the unit nor did it make Brink commanding officer. 
As best he could under those uncertain conditions, Hunter discharged the 
duties of commanding officer, while Brink and he trained the men under Win- 
gate's general supervision. As it developed, circumstances permitted two 
months in which to weld the Galahad volunteers into a homogeneous force. 

In October, theater headquarters listed a block of numbers from which a 
unit designation might be taken. Finally, in late December, Hunter cut the ad- 
ministrative tangle by activating the 5307th Regiment (Provisional), using one 
of the block of numbers supplied by CBI Theater headquarters. 72 

As the unit took shape in India, it comprised three battalions, the 1st, com- 
manded by Lt. Col. William L. Osborne, the 2d, by Lt. Col. George A. McGee, 
Jr., and the 3d, by Lt. Col. Charles E. Beach. Each battalion was broken down 
into two combat teams of 16 officers and 456 enlisted men. The combat team 
had a rifle company of three rifle platoons and a heavy weapons section, a heavy 
weapons platoon to support the rifle company, a pioneer and demolition 
platoon, a reconnaissance platoon (I & R platoon), and a medical detachment. 
The combat team had 306 Ml rifles, 52 submachine guns, 86 carbines, 4 81-mm. 
mortars, 4 60-mm. mortars, 2 heavy machine guns, 2 light machine guns, and 
3 2.56-inch rocket launchers. 

Pack transport was provided for mobility in jungle and over rough terrain. 
Galahad began its campaign with an animal strength of 700. 7 3 

69 Rad AMMDEL 1850, Ferris to Stilwell, 20 Oct 43; Rad WAR 3837, Marshall to Stilwell, 
Ferris, and Wedemeyer, 13 Nov 43. Items 1097, 1257, Bk 4, JWS Personal File. 

70 Ltr, Gen Hull to Hunter, 10 Sep 43, sub: Ltr of Instr. OPD 370.5 CTO (10 Sep 43), 

71 Lt r. Hq Rt E ch USAF CBI to Brink, 13 Nov 43, sub: Outline of Duties. Correspondence 
file citec j n. 6 8(5) 

72 History of CBI, Sec. II, Ch. XIV, Galahad, pp. 5-7. (Hereafter this section of History 
of CBI will be cited as Galahad.) 

73 Galahad, p. 8. 



The ultimate source of Galahad's strategic mobility, its contemplated 
ability to operate against the flanks and rear of the Japanese, was to be air sup- 
ply. So that there might be the highest degree of integration and co-ordination, 
the unit had its own air supply section under Maj. Edward T. Hancock. Its 
duties were to include preplanning airdrops, ready response to requisitions 
from the field, packaging supplies in a manner to permit the safest delivery and 
quickest use, and accurate and speedy delivery of supplies to the point 
prearranged. 74 

There was a certain amount of debate on the proper use of long-range pene- 
tration groups. At the Quebec Conference, General Brooke, Chief of the Im- 
perial General Staff, had expressed the thought that the long-range penetration 
groups should be worked in close co-operation with the units in contact with 
the principal Japanese forces, rather than for dramatic penetrations deep into 
Burma. There was some comment that the long-range penetration groups were 
really wasted in milling about the interior of Burma, that their proper use, 
given the circumstances of jungle terrain and air supply, was for short envelop- 
ments. 75 As events proved, this was Stilwell's idea too. But what use Stilwell 
might make of Galahad was still in the future, in October and November 
1943, and under Brink's and Hunter's guidance Galahad was busy rounding 
itself into shape for the events to come. 

The Kachin Rangers 

Most powerful of the Burman peoples in the path of the projected North 
Burma Campaign were the Kachins. They are a great righting stock who have 
cut their way into Burma from the mountains to the north. Expert woodsmen, 
and uncannily adept at invoking the nats, or minor deities, of the surrounding 
hills, the Kachins reminded some of those Americans who worked with them 
of the American Indian in his greatest days. They had a trait that sometimes 
amused and sometimes touched the Americans who sought to enlist them 
against the Japanese: their culture did not recognize deceit in personal rela- 
tions. "The Japanese have sent me to spy on you," said the Kachin as he 
entered the camp. "Please, how do I begin?" The Kachins' potentialities as 
scouts, guides, and irregulars were obvious, and so Detachment 101 of the Of- 
fice of Strategic Services was organized in 1942. The force thus formed was 
known as the Kachin Rangers. 

Detachment 101 performed a variety of missions, using a mixed personnel 
of Kachins, Burmese, and Americans. An intelligence net was set up in the Jap- 
anese communications zone in north Burma. Guerrilla forces were organized 
around a cadre of trained Americans and Kachins to attack Japanese lines of 
communications, working parties, and patrols, and to identify targets for 
Eastern Air Command. 

74 Galahad, p. 12n, and App. 3. 

75 Min, CCS 107th Mtg, 14 Aug 43. 



A SQUAD OF KACHIN RANGERS prepares for inspection. 

Late in 1942 a training school for Detachment 101 was organized at Nazira, 
Assam, under Capt. William R. Peers. By the middle of 1943, the school was 
able to accommodate 150 students. The training included techniques of espion- 
age and counterespionage, communications, weapons, woodcraft, and Japanese 
organizations, methods, and order of battle. 

After being trained, Burmese and Kachin agents and U.S. officers were 
flown into north Burma behind the Japanese lines, where they proceeded to 
recruit, train, and equip bands of Kachin warriors. Silver rupees, cloth, raw 
opium, and medicines were used as payment. When the time for opening the 
campaign drew near, Detachment 101 had an intelligence net established be- 
hind the Japanese lines, had numerous parties of armed and warlike Kachins 
totaling several hundred, and was steadily expanding its operations. 76 

The Battleground 

Separated from India, China, and Tibet by an inverted U-shaped bend of 
great mountains, the Himalayas and their giant spurs, north Burma is divided 

76 (1) Intervs with Lt Col William R. Peers and Capt James L. Tilly. OCMH. (2) NCAC 
History, App. 8, Brief Sketch of Detachment 101 in the NCAC Campaign. 



by the Kumon Range into two compartments. 7 7 \(Map 2)\ West of the Kumon 
Range, the headwaters of the mighty Chindwin River, flowing north, carved 
out the Hukawng Valley. Directly south of the Hukawng Valley is the ridge 
of Jambu Bum, which separates the watershed of the Chindwin from that of 
the Irrawaddy, Burma's greatest river. South of Jambu Bum lies the long nar- 
row Mogaung River valley, a natural avenue of approach to the great valley 
of the Irrawaddy, which to its inhabitants is Burma proper. East of the Kumon 
Range a tributary stream of the Irrawaddy points a long slim finger north from 
Myitkyina to the Himalayas; at the head of this valley is Fort Hertz, whose 
airstrip and garrison were the last Allied foothold in Burma in 1942 and which 
had been held by the Allies ever since. 

On the western mountain slopes of Burma are tropical rain forests. In the 
Hukawng Valley proper the trees are smaller than the mountain giants, with 
thick, hobbling underbrush about their feet. The occasional clearings more 
likely than not were filled with elephant or kunai grass, tall as a man, and with 
an edge like a samurai sword. Crossing bamboo clumps often involved cutting 
a tunnellike path through the rank growth. The bush was not the tropical 
forest of legend, with rich and exotic fruit growing in profusion. Little to eat 
could be found, while the local people raised only enough rice for their own 

The winter in north Burma is decidedly chill and a distinct ground mist 
often cuts visibility sharply. Noontime is pleasant; then the night air brings 
cold winds with it. The monsoon can come any time after April, and with it, 
floods. The dry-weather road running from north to south down the Hukawng 
and Mogaung valleys has several stretches which are submerged during the 
rains. After February the temperature begins to rise. March is hot and humid, 
and the weather grows progressively worse until the monsoon breaks, usually 
in late May or early June. 

The place names on the map of north Burma might signify as many as a 
hundred bamboo huts, surrounded by a stockade, or might be just a clearing in 
the jungle, like Inkangahtawng, scene of a sharp engagement in the campaign. 
Myitkyina, the metropolis, had about 8,000 people. 

Insect life of a most unpleasant sort is abundant in north Burma. Three 
varieties of leeches lie in wait for their warm-blooded victims, animal and 
human. Some drop from trees when their prey passes below; others, on the 
trailside vegetation, brush off on skin or clothing. There are small black flies 
whose bite is poison, and clouds of buffalo flies that can penetrate any net. The 
ubiquitous malaria-bearing Anopheles mosquito lies in wait, and in the grassy 
clearings lurks a deadly variety of typhus, at that time largely unknown to 
scientist and soldier alike. 

77 This section is based on Merrill's Marauders (Washington, 1945). This popular account is 
one of the AMERICAN FORCES IN ACTION series originally prepared for distribution in 



Planning the North Burma Campaign 

Planning the Chinese Army in India's (CAI) share in the occupation of the 
northern part of Burma was eased because of the great distance between the 
CAI and the principal British concentration which was opposite central Burma. 
The two campaigns would be far enough apart to permit the two command- 
ers considerable freedom even though the operations were ultimately inter- 
dependent. Stilwell's New Delhi headquarters began to plan a north Burma 
campaign in 1942, assuming use of an army corps. After the Trident Confer- 
ence, May 1943, and in compliance with its directive to reoccupy north Burma, 
General Boatner, then commanding the Combat Troops, Ledo Sector, ordered 
his G-3, Col. Robert M. Cannon, to prepare appropriate staff studies. 

Colonel Cannon was directed to prepare a plan to accomplish the current 
mission of protecting Ledo and the Ledo Road project, and the future one of 
clearing the Japanese from North Burma. Boatner desired to make increased 
use of aviation for tactical support and supply, to move the troops by successive 
advances from one dominant terrain feature to another, and to use airborne and 
parachute troops. 

Boatner's directive focused staff attention on the Hukawng Valley, and its 
adjacent terrain compartment, the Taro Plain. After crossing the Patkai Range, 
the Ledo Road emerged from the mountains at Shingbwiyang, at the north- 
western corner of the Hukawng Valley. About ten miles south of Shingbwiyang 
is the Tanai Hka, whose course, though winding snakeiike, still follows a 
definite direction from southeast to northwest. So the road's trace would initially 
run roughly parallel with the Tanai. The traveler who wishes to cross the Tanai 
near Shingbwiyang does so at the Kantau ford. Since at the head of the Hu- 
kawng Valley the Ledo Road would run roughly parallel with the Tanai, the 
latter's tributaries would offer the first water obstacle to the road's progress. 
The first tributary the road builders would meet would be the Tarung Hka. 

The western wall of the Hukawng Valley is the Wantuk Bum. Over the 
eons of geologic time, the Tanai has broken through the Wantuk Bum south 
of Shingbwiyang. The river bends sharply, flowing almost due south as it 
enters the small valley of the Taro Plain. The Taro Plain, therefore, is like a 
small closet adjacent to the long narrow room of the Hukawng and Mogaung 
valleys. He who wishes to enter the long narrow room from its northern door 
must be sure that no one is lurking in the closet. 

Colonel Cannon complied with Boatner's directive by preparing a three- 
phase plan calling for seizure of the line of the Tarung Hka from Sharaw Ga 
(ga indicates village) to the confluence of the Tarung Hka and Tanai Hka. The 
advance would begin either when the roadhead reached the village of Namlip 
Sakan or when ordered. The 38th Division, supplied by airdrop which would 
be supplemented by pack train and porter, would be used. When the roadhead 
reached Chinglow Sakan, assumed to be 1 January 1944, the next phase would 

MAP 3 



begin, seizure of the Jambu Bum ridge line, which, running east and west, sep- 
arated the Hukawng and Mogaung valleys. The 2 2d Division would join the 
38th Division. To carry the advance beyond the Jambu Bum, Colonel Cannon 
suggested using Chinese airborne troops against either Myitkyina or Kamaing 
and urged that all supply be by air. 78 Approved by Stilwell, these preliminary 
studies incorporated the elements of air supply and advance from terrain fea- 
ture to terrain feature that marked the final plan. 79 

For the drive on Myitkyina, Stilwell's planners drafted Albacore , a devel- 

opment and refinement of the earlier Cannon suggestions. (Map 3) Albacore 
One and Albacore Two provided for protecting the Ledo base and securing 
the Shingbwiyang area on D minus 15, in that order. Albacore Three un- 
folded in four phases, 3 A, 3B, 3C, and 3D. Phase 3 A was to seize the Jambu 
Bum; 3B, the Lonkin-Kamaing line; 3C, Mogaung and Myitkyina; 3D, Katha 
and Bhamo. Phase 3A called for the 38th Division to advance from the Tarung 
Hka line on D Day, or on 1 December 1943 if D Day had not been announced 
by then. One regiment was to drive up the Tanai valley, adjacent to and east of 
the Hukawng Valley, another was to take the line of the Nambyu River and 
go south and occupy the Jambu Bum. The third regiment was held in reserve. 

When the 38th was near the Jambu Bum, the 22d Division would move 
into the Shingbwiyang area. One of its regiments would protect the 38th Divi- 
sion's right flank, the other would fly to Fort Hertz. This latter move bore the 
waggish code name Ledo Striptease. Phase 3B called for the 38th Division 
to advance from a line just below the Jambu Bum in two columns (one up the 
Tanai Hka valley, the other along the dirt road to Kamaing). The 22d Division 
would be in forward reserve. 

In phase 3C the 38th Division would take Mogaung. The 2 2d Division 
would follow in column until Mogaung was taken, then swing around the 
Kumon Range to attack Myitkyina from the south while the regiment earlier 
sent to Fort Hertz closed in from the north. In Phase 3D the 38th Division 
would take Katha, the 22d Division, Bhamo. Albacore assumed the Japanese 
were not in strength north of Kamaing. 80 

Stilwell, surveying the field as his staff put the final touches on the Chinese 
and American share of the campaign, was quietly confident in a personal radio 
to Marshall. Stilwell was satisfied with the Chinese Army in India, which was 
well trained and in good condition. If the Japanese did not reinforce materially 

78 (1) Draft Plan, sgd Col Robert M. Cannon, ACofS, G-3 [probably May 43], s ub: P lans 
for Offensive Opns Employing Para and Airborne Trs. NCAC Files, KCRC. (2) See lCh. Ill, I 

79 Memo, CG, Combat Trs, Ledo Sector, for ACofS, G-3, 26 May 43, sub: Tactical ans; 
Ltr, Cannon to CO, Base Sec No. 3, 12 Jun 43, sub: Tactical Plans. NCAC Files, KCRC. 

80 (1) Opn Plan Albacore, 8 Aug 43. Folder, Albacore, Ledo Striptease, NCAC Files, 
KCRC. (2) Growth, Development and Operating Procedures of Air Supply and Evacuation Sys- 
tem, NCAC Comd Front, Burma Campaign, 1943-45, prep by U.S. Mil Observers' Gp, New 
Delhi, India, MS (hereafter, NCAC Air Supply), p. 8-1. The authors' pagination form has 
been retained in citations of this manuscript throughout the present volume. OCMH. 



in the Myitkyina-Pao-shan area, the operation had a good chance of success, 
and if 4 Corps did its share by driving into Burma from Manipur State, the Jap- 
anese could not reinforce. He planned to take Shingbwiyang and build an air- 
strip there from which to supply one division to push toward Mogaung. When 
the Ledo Road reached Shingbwiyang, two divisions driving south could be 
supplied by air. With any luck, Stilwell expected that the Ledo Road would 
reach Shingbwiyang by 31 December. 81 

Underlying these plans and hopes was Chih Hui Pu's estimate of Japanese 
strength and dispositions. Chih Hui Pu believed that the Japanese in Burma 
were commanded by Headquarters, Burma Area Army, at Rangoon, with field 
command by Headquarters, 15th Army, at Maymyo, Burma. Under 15th Army 
were four divisions, the 18th (Mandalay), 33d (Monywa), 55th (Akyab), and 
56th (Lashio). Four Thai divisions were thought to be in the Shan States of 
Burma. There were reports, as yet unverified, that four more Japanese divisions, 
identified as the 14th, 16th, 23d, and 2 1st were in Burma. As of 30 October 
1943, G-2 of Chih Hui Pu wrote: "There is no evidence of any enemy in 
strength north of the fortified area of Kamaing nor is there any indication of 
such a move in prospect. There are however troops available south of 
Kamaing which can be moved up when and if needed." Adding faith to this 
estimate were the circumstances of complete U.S. air superiority over the Hu- 
kawng-Mogaung valleys, which permitted unrestricted aerial reconnaissance, 
and the host of friendly Kachin informants. 82 

Actually, Chih Hui Pu erred. The Japanese Burma Area Army had six divi- 
sions, the 18th, 31st, 33d, 54th, 55th, and 56th, of which the 31st and 54th ar- 
rived in 1943, and the 24th Independent Mixed Brigade, then being organized. 
The 54th Division went to garrison the Arakan region against the feared am- 
phibious descent by SEAC; the 31st Division, to central Burma opposite Mani- 
pur State in India. 83 ( Chart 4 ) The Thai divisions may be disregarded for they 
never met SEAC troops in battle and did not free any Japanese garrison troops 
for service elsewhere, as the Japanese in Burma were not faced with civil unrest. 
And the Japanese were north of Kamaing. 

Appraising Albacore, General Boatner thought the scheme had flaws. 
Though he knew of no Japanese north of Kamaing, he feared that a forward 
displacement to the line of the Tanai followed by a pause might well attract 
the Japanese, so he suggested holding the advance well north of Shingbwiyang 
until 15 November and then going all out for the Jambu Bum. Stilwell agreed 
and was willing to hold the advance until the 15 th. 

81 Rad CK 36, Stilwell to Marshall, 30 Aug 43. Item 769, Bk 2, JWS Personal File. 

82 (1) Wkly G-2 Rpt, Hq Chih Hui Pu, 30 Oct 43. KCRC. (2) Japanese Officers' Comments, 
Incl 4. OCMH. Lt. Gen. Shinichi Tanaka states that the Japanese order of battle did not include a 
14th, 16th, or 23d Division. (3) A Japanese area army may be considered the equivalent of an 
American army and a Japanese army the equivalent of an American army corps. Therefore on 
the maps the symbols used for enemy units are those of the American equivalents. 

83 (1) Japanese Study 89, pp. 9-10, 30. (2) Japanese Officers' Comments, pp. 2-3. OCMH. 



Chart 4 — Japanese Organization and Dispositions 
November 1943 

'Spina ni^3P r(d 

h Airltinf in Moloyo 
'En icu't In Burma. 
■£n null le >>it AmkaA. 
So«r« )rjpa»<Sc Off jfW Commfnh, p S 

Brig. Gen. Benjamin G. Ferris, chief of staff of StilwelPs Rear Echelon the- 
ater headquarters at New Delhi, objected to General Boatner's proposal. Ferris 
stated that the principal reason for the move forward was to occupy ground 
within which to build an airstrip in order to receive road-building machinery, 
which would be flown in to start construction back to the current roadhead. If 
the Ledo Road did not reach Shingbwiyang by January 1944, said Ferris, all 
hope of a link with China in 1944 was gone. Further, an advance would set the 
precedent for a similar forward displacement by 4 Corps, whereas postpone- 
ment would be highly contagious. In summarizing, General Ferris gave first 
place to the argument that the advance would give Stilwell a potent argument 
in urging an aggressive attitude on SEAC and the Combined Chiefs of Staff. 
Stilwell initially agreed with Boatner, but he then discussed the matter per- 
sonally with Ferris and the latter persuaded him to let the orders stand. Boatner 
fared no better with a suggestion that the entire 38th Division be sent forward. 
Brig. Gen. Frank D. Merrill tried to arrange it, but Stilwell could not agree, 
because the supply aircraft to support such a move were not at hand. 84 

84 (1) Rad RE LOT G 240, Boatner to Stilwell, 4 Oct 43; Rad AD 2213, Stilwell to Boatner, 
6 Oct 43; Rad AMMDEL 1728, Ferris to Stilwell, 6 Oct 43; Rad AD 2223, Stilwell to Ferris, 7 Oct 
43; Rad B 134, Stilwell to Boatner, 6 Oct 43. Items 1062, 1060, 1064, 1065, 1063, Bk 4, JWS 
Personal File. (2) Ltr, Frank [Merrill] to Haydon [Boatner], reed 4 Sep 43. Folder, Combat Rpts 
FE and Ln O's, Albacore Hist File, NCAC Files, KCRC. 

MAP 4 



The Campaign Begins 

Albacore Two called for a forward displacement of one regiment of the 
Chinese 38th Division to cover the onward movement of the Ledo Road and 
to reach the Tarung, which was the line of departure for the projected North 
Burma Campaign. \(Map 4) | D Day, if previously unannounced, was to be 1 
December 1943. The orders as received by General Sun for his 38th Division 
left him very little scope for his own initiative. He was directed to send the 
112th Regiment forward to shield the advancing road builders. The 2/1 12th 
was to occupy the Tarung Hka villages of Sharaw Ga and Ningbyen, the 
1/1 12th to occupy Yupbang Ga, also on the Tarung, and the 3/112th, Ngajat- 
zup at the northern edge of the Taro Plain. 85 This deployment, which dispersed 
the regiment, was designed to hold the line of the Tarung and the Tanai by 
controlling the fords, and to bar Japanese excursions from the Taro Plain. Chih 
Hui Pu expected that the 112th could readily brush aside the maximum ex- 
pected resistance which was assessed as scattered parties of Burmans under Jap- 
anese leadership. 86 

General Sun received his orders on 5 October but hesitated over moving 
forward. Boatner saw no good reason for Sun to delay, saying that air support, 
both tactical and supply, was ample, and enemy opposition "certainly insignif- 
icant." 87 Finally, the Chinese battalions began to move out. As they trudged 
ahead, they passed over portions of the trail by which the pathetic refugees of 
1942 had fled Burma. The path was a ghastly sight; skeletons were seen about 
every water hole and at the beginning of every slope. Huddled groups of bones 
showed where small refugee camps had perished en masse. On his way back to 
Burma, Dr. Gordon S. Seagrave, the Burma Surgeon of the first campaign, saw 
"hundreds and hundreds" of skeletons as his colleague, Col. John M. Tamraz, 
the SOS surgeon, had eight months before. 88 

Ordered by Chih Hui Pu to occupy Sharaw Ga and Ningbyen, the 2/1 12th, 
38th Division, was a little task force as it moved deeper into Burma. It was 
reinforced and supported by the 5th Company of the 114th Regiment, a Sea- 
grave hospital unit, engineer, and quartermaster troops. 89 Two miles north of 

85 For economy of space and simplicity, a battalion is sometimes identified by giving first its 
numerical designation, followed by a slash, and then the name or the number of the parent 
regiment, as in the example above, the 2/1 12th, or the 2d Battalion of the 112th Regiment. 

86 (1) NCAC History, pp. 25-28. (2) Lt Col Thomas F. Van Natta, History of the 38th 
Div, CAI, 1 Oct 43-31 Aug 45, MS. NCAC Files, KCRC. (3) Ltr, Boatner to Stilwell, 29 Nov 
43. Folder, Combat Rpts FE and Ln O's Albacore Hist File, NCAC Files, KCRC. 

87 Rad RELOT G 257, Boatner to Stilwell, 15 Oct 43. Item 1085, Bk 4, JWS Personal File. 

88 (1) Tamraz Diary, 19 Feb 43. SGO Hist Div, Washington, D.C. (2) Dr. Gordon S. 
Seagrave, Burma Surgeon Returns (New York: W. W. Norton & Co., Inc., 1946), pp. 27-28, 
72-73, 60-63. 

89 Regimental designations of Chinese infantry divisions were obtained by multiplying the 
division's number by three. The product designates the last regiment of the division. Thus, the 
three regiments of the 38th Division were the 112th, 113th, and 114th; of the 22d Division, 
the 64th, 65th, and 66th Regiments. 



Sharaw Ga and many weeks before D Day, the Chinese at about 1200, 30 
October 1943, met an enemy outpost.The outpost was driven back, and the ad- 
vance guard moved on to the Sharaw Ga clearing, where mortar and machine 
gun fire held it until dark. On the next day the remainder of the battalion came 
up, and the Chinese tried to take the village, which lay between two hills. 

The hill to the north was lightly held. The Chinese quickly overran it, then 
came under such heavy fire from the second hill that they had 116 casualties. 
On 1, 2, and 3 November the Chinese attacked, achieving nothing and losing 
fifty more men. Then they went on the defensive and dug in, but managed to 
patrol and keep in touch with the regimental command post at Ningam Sakan. 
The 1st Battalion had much the same experience at Yupbang Ga. Encountering 
a well-entrenched and well-led force, it dug in and was quick ly isolated by a 
roadblock placed between Sharaw Ga and Ningbyen. \(Map 7j| That left the 3d 
Battalion, which was similarly stalemated on the northern edge of the Taro 
Plain. 90 This well-led, well-entrenched enemy was not the expected rabble of 
Burman levies, but elements of the Japanese 3 6th Regiment, 18th Division, under 
Lt. Gen. Shinichi Tanaka. 

The surprise caused by the presence of these excellent veteran troops was 
complete. G-2 reports had given no hint that Japanese forces were in the area. 
Compounding the intelligence and reconnaissance failure was the reluctance 
of Chih Hui Pu to admit that the enemy was present in strength. As late as 
20 November the weekly G-2 report dismissed the Japanese as having seventy- 
five to one hundred men at Yupbang Ga, with one hundred more pocketed 
between Sharaw Ga and Ningbyen, and added that judging by road traffic a 
battalion was moving up to reinforce. 91 

Drawn from the island of Kyushu, the men of the 18th Division considered 
themselves authentic heirs of the martial traditions of the hot-blooded Kyushu 
clans. A diploma of honor had been given them for their pre-eminent share in 
storming the island of Singapore. Their commander, General Tanaka, had been 
chief of operations of Staff Headquarters in Tokyo and had gone from there to 
join the staff of Southern Army. From Southern Army he had taken command of 
the 18th Division when Lt. Gen. Renya Mutaguchi moved up to command of 
the 13 th Army. 

The 18th Division had moved forward to carry out its mission of garrison- 
ing north Burma, to which was added in September that of supporting a 
planned attack on India. Anxious to prevent interference with its offensive, the 
Japanese headquarters in Burma, Burma Area Army, was improving its posi- 
tions all around the perimeter of Burma. The 36th Division in October 1943 
eliminated a Chinese bridgehead over the Sal ween, north of Teng-chung. In 
late September, 13th Army in anticipation of the dry season ordered 18th Divi- 
sion to carry out, with its main force, a delaying action in the Hukawng Valley 

90 (1) Map Overlay, Hq 5303d (Prov) Combat Trs, 13 Nov 43. NCAC Files, KCRC. (2) 
NCAC History, pp. 25-28. 

91 Wkly G-2 Rpts, Hq 5303d Combat Trs, Oct-Nov 43; Numbered Int Sums, summer and 
early fall 1943, Hq Combat Trs, Folder, Int Sums, NCAC Files, KCRC. 



against the expected Allied thrust from Ledo. Myitkyina was to be strongly 
held to block any attack from China. 

About 24 October a company of the 2/ 56th, 18th Division, had arrived in 
the Tarung-Tanai area on reconnaissance. Not finding any Chinese immediately 
to the north of them, they constructed defensive posts in and around Ningbyen. 
Probably at the first clash the Japanese outposts appealed for help. A Japanese 
concentration of unknown size was at Maingkwan and the remainder of the 
2/ 56th was sent forward from there "in the earliest part of November." The 
rest of the 18th Division was "a month's march" away, General Tanaka recalled 
in 1948. 92 It may be assumed that the 18th Division was reinforced as quickly 
as possible and was soon present in a strength well able to meet the 112th 
Regiment's three scattered battalions on more than even terms, especially since 
at first the Japanese tactics were to place blocks across the trails and force the 
Chinese to attack them. 

The Tarung, which flows south through this area to enter the Tanai, is a 
respectable river, two hundred yards wide in the dry season and much wider 
during the rains. Firmly entrenched at Sharaw Ga and Yupbang Ga, the Jap- 
anese were holding the river crossings and thus the springboards for the of- 
fensive General Tanaka was speedily planning. 93 

Tanaka personally reconnoitered the area in early November. He decided 
to adopt a plan which he described long after as being "to move the main 
strength of the division from Ningbyen toward Shingbwiyang and the exit 
of the mountain road on the India-Burma border to attack and destroy the 
American and Chinese forces which would advance in a long column through 
the tortuous Ledo Road in India." The operation was to begin on 15 December 
1943. 94 

As the Japanese grew in strength, they became more aggressive, and the 
112th Regiment, 38th Division, had a series of misfortunes in November. One 
of its companies was annihilated on 2 November. The regimental command 
post was overrun on the 3d as its guard was digging in for the night. The regi- 
mental commander, a Colonel Chen, and the junior U.S. liaison officer, Maj. 
George T. Laughlin, escaped, but the chief liaison officer, Lt. Col. Douglas G. 
Gilbert, was captured. A company of the l/ll4th, rushed up to aid by reopen- 
ing the trail to the 1/1 12th near Yupbang Ga, was halted by the Japanese well 
short of its goal. Air supply brought rations to the besieged 150 survivors of 
the 1/1 12th, but the acute water supply problem had to be met in part by tap- 
ping jungle vines. A great banyan tree was ingeniously made into a fort defend- 

92 (1) Rpts cited n. 91, above. (2) Interrog, Gen Tanaka, 13 Jan 48. OCMH. (3) SEATIC 
Bull 244, 3 Oct 46, sub: History of Japanese 33d Army, pp. 11-12. MID Library. (4) The 
diary of a Japanese officer indicates that he was in Taihpa Ga with his unit before 24 October 
1943. He recorded that incoming soldiers were digging defensive positions, presumably on 25 
or 26 October. General Tanaka thought that the Japanese arrived about 26 October. Folder, 
Misc Work Sheets of Captured Japanese Docs— 1944; Int Sum 111, 6 Nov 43, Hq 5303d (Prov) 
Combat Trs. Folder Int Sums, NCAC Files, KCRC. (5) Japanese Officers' Comments, p. 5. 

93 Interrog cited n. 92(2). 

94 Japanese Officers' Comments, Incl 4. OCMH. 



ing the west end of the dropping field and garrisoned by two squads with 
heavy machine guns who put themselves in, around, and under the mighty 
tree. 95 

The 38th's failure to advance brought vigorous exchanges between Generals 
Sun and Boatner. Boatner believed there was only one Japanese battalion at 
hand, that by passivity and bad tactics the Chinese had let themselves be sur- 
rounded and then proceeded to waste their ammunition. Sun thought the 
whole 36th Regiment faced him along the Tarung and wanted reinforcements. 
On 26 November Boatner told Maj. Gen. Thomas G. Hearn, Chief of Staff, 
U.S. Army Forces, CBI, that the situation was "critical," that Sun wanted to 
retreat, and asked if General Hearn could intervene. Since Stilwell was absent 
on a highly secret and most important mission, Hearn had to handle the prob- 
lem, which he did by telling Boatner to act strongly in Stilwell's name. 96 

One possible difficulty lay in the fact that Chih Hui Pu's orders to Sun had, 
as noted before, allowed him very little initiative and even prescribed the loca- 
tion of his several battalions. Moreover, he had not been free to reinforce with- 
out Chih Hui Pu's approval. The situation slowly improved when Chih Hui 
Pu began reinforcing the 112th Regiment, 38th Division. The 114th arrived 
at the front in mid-November, and the remainder of the division was on the 
way. 97 


As the campaigning season of 1943-44 began, the most important activity 
took place on the lower levels of the theater. The problem of improving the 
Assam line of communications to an acceptable standard was approached with 
determination and vigor. In far-off Washington, the President introduced a 
new element by his decision to send the B-29's to CBI. Then came the en- 
counter between Chinese and Japanese in the Hukawng Valley which upset the 
timetable for the campaign. D Day was to have been in December, leaving 
time for SEAC to settle on a plan for the campaign which would provide a di- 
rective to the Chinese in north Burma and the British forces in Manipur and 
the Arakan. In accord with that concept, the Chinese in north Burma moved 
forward to screen the advance of the Ledo Road, well before D Day. They 
stumbled on a strong force of Japanese, and on 30 October 1943 the fight 
began with both sides reinforcing. The Burma campaign was under way though 
the detailed plan for it was not yet approved by the CCS and the basic prepara- 
tion, improvement of the Assam line of communications, had just begun. 

95 Comments by Col Dupuy on NCAC History, I, 29. (Hereafter Dupuy Comments.) OCMH. 
Colonel Dupuy was present throughout the North Burma Campaign and kept extensive notes. In 
writing this section, the authors are greatly indebted to him and Lt Col George T. Laughlin.- 

96 Rad RELOT G 323, Boatner to Hearn, 26 Nov 43; Rad RELOT G 322, Boatner to Hearn, 
26 Nov 43; Rad OT 56, Hearn to Boatner, 30 Nov 43. Items 1286, 1285, 1290, Bk 5, JWS 
Personal File. | ^^_^^_ 

97 (1) Map Overlay cited | n. 90(1)1 (2) Map Overlay, Hq 5303d (Prov) Combat Trs, 4 
Dec 43. NCAC Files, KCRC ^ 


Sextant: The Watershed 

At the Casablanca, Trident, and Quadrant Conferences a strategy whose 
successful execution would break the blockade of China had been roughly 
shaped. Pledges had been given to the Chinese, notably that of Trident: "No 
limits, except those imposed by time and circumstance, will be placed on the 
above operations, which have for their object the relief of the siege of China." 
This statement had followed on a Chinese threat to seek a separate peace. 
There was another question: how long could China survive blockade? Stilwell, 
Chennault, the President, the Prime Minister, all agreed at Trident that 
China must have aid soon. Another powerful influence in shaping Allied 
strategy had been the President's wish that China be treated as a Great Power, 
that it join in the councils of the Great Powers as an equal. 

To complete Allied plans for the relief of China, the President arranged for 
the Generalissimo to meet with him and Mr. Churchill at Cairo in November 
1943. Then and there the threads would be drawn together. The Generalissimo 
would confer with his colleagues; the final details would be added to the plan 
for China's relief; the dignitaries would approve it; and a CCS directive to 
SEAC would be issued. The Cairo Conference was the high point, the water- 
shed, that divided Sino- American relations. After Cairo, the currents flowed in 
a very different direction. 

Drafting SEA Cs Proposals 

When Admiral Mountbatten on 1 November 1943 formally opened his 
headquarters as Supreme Allied Commander, Southeast Asia Command, one of 
his first tasks was to prepare a plan to submit to his superiors, who if approving 
it would provide the necessary additional resources, landing craft, principally. 
To play his part in this planning was Stilwell's first duty on leaving Chungking 
for India in October 1943. Independently, Mountbatten and Stilwell had come 
to similar conclusions on the preliminary studies prepared by General Head- 
quarters (India) in the last phases of that body's concern with Burma opera- 
tions. When General Auchinleck in September 1943 proffered a plan calling for 
the now-familiar converging attacks on Burma from Yunnan, Ledo, Assam, and 
the Arakan, Stilwell had been critical. The scheduling of the proposed several 
drives upset him, for he found them so separate in time as to open the prospect 



of the Allies' being defeated in detail. And he added: "I understood the orders 
to call for Vigorous and aggressive action' and I don't find a hell of a lot of it 
in the plan. However, we will proceed as indicated and perhaps our doubts will 
be resolved when Admiral Mountbatten arrives." 1 

After examining the same plans Mountbatten, too, was critical, but where 
Stilwell was characteristically blunt, Mountbatten was urbane: "There is also 
no doubt that the climate and the antiquated and close methods used in India 
have their effect on the keenness of officers after a year or two and so I have 
found that the plans made by the Indian Staff are somewhat pessimistic and 
unenterprising." 2 

When the work of preparing SEAC's proposals to the CCS, the President, 
the Prime Minister, and the Generalissimo began, Stilwell submitted his views 
as did General Headquarters (India) and the local combined planners. 3 

The proposals and decisions that began to form fell into two categories, 
those for the first phases which SEAC could execute with its own resources, 
and those which needed approval and support by higher authority. Almost 
immediately Stilwell received his orders and approval of the opening phases of 
Albacore Three, which called for him to establish a bridgehead over the 
Tanai. As for the Arakan, the SEAC minutes state that General Giffard was not 
satisfied with the safety of Chittagong while his troops held their current posi- 
tions, so he proposed that t hey make a twenty-mile advance to secure the 
Buthidaung-Maungdaw road ] f See Map 1 \ -inside back cover.) 

Neither Admiral Mountbatten nor General Slim was content with this 
modest contribution. General Giffard according to the minutes "agreed that it 
was mainly a defensive move." So Giffard's orders were changed to call for the 
exploitation of any success, with Akyab the objective. It was further agreed in 
SEAC that whatever the ultimate objective in Burma, 4 Corps, on the central 
front, at some time would be obliged to advance through the noxious and 
malarial Kabaw Valley. The Arakan advance might begin the second week of 
January 1944 ; 4 Corps and the Chinese forces in Yunnan (Y-Force) would 
move out in early March; and any airborne operation would be in mid-March. 4 
It is notable that Stilwell was thus directed to advance into hostile territory as 
part of a larger operation whose objectives had not been defined and whose 
resources were not at hand. However, none doubted that all would be provided 
in due time. 

Of the three courses seen as open to 4 Corps and the air-supplied light 
infantry of the British Long-Range Penetration Groups (LRPG's or Chindits), 

1 (1) Ltr, Auchinleck to Stilwell, 7 Sep 43; Quotation from Ltr, Stilwell to Auchinleck, 16 Sep 
43. Item 226, Bk 3, JWS Personal File. (2) Mountbatten Report, Pt. A, par. 7. 
1 Extract, SAC's Personal Diary, 30 Oct 43. SEAC War Diary. 

3 Stilwell's planhasnot been found. Probably it is reflected in the views expressed in the mem- 
orandum cited ir fnote7l below. 

4 (1) SEAC Plan, SAC (43) 2, 28 Oct 43; Min, SACs Mtg, 31 Oct 43; Rad 7, SEACOS to 
COS, 31 Oct 43. SEAC War Diary. (2) Rad AMMDEL 1963, Merrill to Stilwell, 3 Nov 43. Item 
1 162, Bk 4, JWS Personal File. 



the SEAC planners under General Wedemeyer, SEAC's deputy chief of staff, 
preferred Toreador, an airborne landing by two divisions in central Burma. 
If successful in the operation's first phases, the divisions would exploit toward 
Mandalay. The other alternatives considered were: (1) an overland advance 
toward Ye-u; (2) Tarzan, an airborne landing at Indaw on the railway to 
Myitkyina, a drive by the Ledo force on Myitkyina, and a bridgehead over the 
Chindwin to be established by 4 Corps. 

Observing the drift of the planning, Stilwell grew concerned and prepared 
a critique on 27 October 1943 which he submitted on 3 November. 5 The 
critique stated that all SEAC plans to date had been closely based on estimates 
of the logistic situation, that they had been "permeated by fear of failure or 
reluctance to take the bold course." Singling out Tarzan, which Mountbatten 
appeared to prefer, Stilwell remarked (giving an incorrect figure) that the oper- 
ation comprised 80,000 troops in the Arakan, 6 limited to advancing on Akyab, 
with nothing further contemplated; an advance from Imphal to the Chindwin 
River; placing a division on the railway to Myitkyina and leaving it there; an 
amphibious operation against the Andaman Islands, which he thought had "no 
immediate bearing on the main problem"; Stilwell's Chinese forces, "left to 
their own resources to effect a junction and open the Burma Road." 

Stilwell believed that if Akyab was taken, this victory should be exploited 
by a series of amphibious hooks down the coast aimed at the port of Bassein. 
Success would give bases from which to dominate Japanese aviation in south 
Burma. In central Burma, he suggested a two-pronged operation aimed at 
Mandalay. "The Indaw operations should be cancelled." His Chinese forces 
would do their best in conjunction with the above. Stilwell stated: 

With the large air and naval units to be committed, nothing less than the above is justified. 
Nothing less than this is either bold or aggressive. Nothing less takes complete advantage 
of our position for concentric attacks. Nothing less threatens the enemy with serious loss. 

I take exception to any trend in the planning which fails to use to advantage our 
overwhelming strength, to any tendency towards vagueness in objectives, to any move which 
does not absolutely require a strong enemy reaction to check it. 

Under present plans, Burma could [Stilwell's italics] be ready to fall to a vigorous attack, 
and for lack of trying, we might not even find this out. In other words, we are not even 
making a reconnaissance in force, let alone a serious attack. 7 

Tarzan, the plan for SEAC's share in the campaign, was nevertheless 
adopted by SEAC on 7 November. Behind the decision lay Mountbatten's 
announced desire for a guaranteed victory, his admission that he would choose 
the less desirable course if it promised success. Tarzan was urged by General 
Headquarters (India) and by Mountbatten's three commanders in chief, 
General Giffard, Admiral Sir James Somerville, and Air Chief Marshal Sir 

5 Stilwell Diary, 3 Nov 43. (See Bibliographical Note.) 

6 Actually present were the 5th and 7th Indian Divisions with three brigades each, and the 
81st West African Division with two brigades. Mountbatten Report, Pt. B, pars. 34-37. 

7 Memo, Stilwell for SEAC, 27 Oct 43. SNF 215. 



Richard Peirse. General Wedemeyer commented that Tarzan would not 
accomplish the objectives given to SEAC in its directive. Representatives of 
CBI Theater headquarters at SEAC were equally critical. Even if all went well, 
they remarked, the monsoon rains would find SEAC having only a bridgehead 
over the Chindwin and an airborne division mired on the railway to Myitkyina. 
This seemed little to show for a season's fighting. The only consolations were 
that the dropping of an airborne division in Burma might open opportunities 
and that SEAC agreed to study operations in Burma to follow Tarzan. From 
the discussions that accompanied the adoption of Tarzan, CBI Theater liaison 
personnel received the impression that Giffard, Somerville, and Peirse were not 
aggressively inclined, placed no value on operations in Burma, and had staffs 
who were too impressed by logistical difficulties and indifferent to what might 
be done to improve the logistical situation. But, Tarzan it was, and the SEAC 
secretariat began to prepare the papers on it and on the over- all plan for Burma, 
now called Champion, for submission to the Comb ined Chiefs of Staff, the 
President, the Prime Minister, and the Generalissimo. 8 p5>g Map 2.)"\ 

The United States Prepares for the SEXTANT Conference 

The President's diplomatic preparations for a meeting with the Generalis- 
simo, the Prime Minister, and the Combined Chiefs of Staff had been under 
way since the Trident Conference in Washington, May 1943. In June Mr. 
Roosevelt told the Generalissimo of his anxiety to meet him, and a discussion 
of times and places followed at once. Originally, the meeting was to have been 
of just the two statesmen, and the Generalissimo suggested Alaska in August 
or September. The course of events made the President feel ever more strongly 
that he should meet with Marshal Joseph V. Stalin and, of course, Mr. Churchill, 
and so the President began to consider co-ordinating the two meetings. 

The foreign ministers' conference at Moscow in October marked further 
progress toward the President's goal of having China accepted as a Great Power, 
for Great Britain and the Soviet Union agreed to China's signing a Four Power 
Declaration. This agreement greatly pleased Roosevelt, who told the General- 
issimo that the ice had been broken, that he and the Chinese statesman had 
now established the principle of China's Great Power position. To arrange a 
meeting between the several statesmen remained, and from this innocent cir- 
cumstance the Chinese insistence on making or keeping face under any and all 
conditions led to great consequences. In so many words, the Generalissimo 
insisted on seeing Mr. Roosevelt before the latter saw Marshal Stalin, or else 
postponing the meeting indefinitely. Roosevelt agreed, and Brig. Gen. Patrick 

8 (1) Rad 22, Wedemeyer to Marshall for Maj Gen Thomas T. Handy, 6 Nov 43; Rad SEACOS 
83834, Mountbatten to COS, 10 Nov 43; Extract, SAC's Personal Diary, 7 Nov 43. SEAC War 
Diary. (2) Rad AMMDEL 2008, Merrill to Stilwell, 8 Nov 43; Rad AMMDEL 2023, Merrill to 
Stilwell, 10 Nov 43; Rad AMMDEL 2036, Merrill to Stilwell 11 Nov 43. Items 1203, 1225, 1236, 
Bk 4, JWS Personal File. 



J. Hurley, who acted as the President's personal representative in the Middle 
East, was sent to Chungking to arrange the details. 9 Thus, the Generalissimo 
sacrificed the strategic advantage of having the last word with the President. 

Even as the next meeting (Sextant) of the Allied statesmen was being 
convened, significant trends in U.S. strategy were depreciating China's impor- 
tance as an ally against Japan. The increasing strength of the U.S. Navy's fast 
carrier task forces and the realization of the B-29's potentialities were leading 
the lower echelons of U.S. planners to an awareness that Japan could be 
defeated without a major U.S. land campaign in China. In summer 1943 the 
Joint Chiefs of Staff decided to use the fast carrier task forces and amphibious 
troops against the Japanese positions in the Gilbert and Marshall Islands. The 
Gilberts were to be attacked in November 1943, the Marshalls, in January 
1944. 10 The decision to initiate action in the Central Pacific did not, of course, 
by itself change China's role in the evolving strategy of the United States, but 
the more the fast carrier task forces prospered in their advance across the Pacific, 
the more islands that fell into U.S. possession, the less need there would be to 
seek the Generalissimo's co-operation. The means for a major thrust across the 
Central Pacific were coming to hand and so was the realization of China's 
diminishing strategic importance. 

The Operations Division observed: 

Despite the agreements that the United Nations should direct their principal offensive 
efforts against Germany and contain the Japanese by a series of relatively minor thrusts, it is 
becoming increasingly apparent that operations against the Japanese are approaching major 
proportions. Plans for the defeat of Japan are not yet firm. However, the degree of success 
enjoyed thus far is indicative of the need of a short-term plan for operations against Japan 
"upon Germany's defeat" with principal emphasis on approach from the Pacific rather than 
from the Asiatic mainland. 1 1 

The Quadrant Conference, Quebec, August 1943, ordered the combined 
staffs to prepare a "short plan for the defeat of Japan." The planners complied 
on 25 October 1943. They suggested four broad possible courses of action, all 
of them bypassing the mainland of China. For operations in China, the Com- 
bined (i. e., Anglo-American) Staff Planners suggested only an eventual limited 
B-29 offensive supported through north Burma by a line of communications 
that would also be called on to support the Fourteenth Air Force and the 
re-equipping of the Chinese Army. 

Of the four proposed courses, the recommended one included taking 

9 ( 1 ) Msg, Roosevelt to Chiang, 30 Jun 43; Ltr, Soong to Hopkins, 21 Jul 43. Bk IX, Hopkins 
Papers. (2) Memo, Hearn for Generalissimo, 1 Nov 43. Item 1139, Bk 4, JWS Personal File. (3) 
Incl to Memo, Somervell to Hopkins, 5 Nov 43. Bk VII, Hopkins Papers. (4) Robert E. Sher- 
wood, Roosevelt and Hopkins, MS, IX-2-93. This manuscript quotes the President as aware of 
China's weakness but as desiring to be friendly with her 400,000,000 people and so wanting 
China to sign the Four Power Declaration. 

10 Min, JCS 97th Mtg, 20 Jul 43. 

1 1 Compilation of Background Material for Sextant, Table 4g, prep by Strategy and Policy Gp, 
OPD. ABC 337 (18 Oct 43) Sec 5, A48-224. 



Formosa in spring 1945, while retaining the option of taking Sumatra in spring 
or autumn 1945 if the Formosa operation had to be postponed. The planners 
concluded there was no prospect of defeating Japan by October 1945. The 
Central Pacific course of action included capture of the Marshall, Caroline 
(Truk area), Palau, and Mariana Islands. If Truk was bypassed, the advance 
might reach the Marianas in July 1944; Truk, in November 1944; and the 
Palaus, by early 1945. It was recognized that good bomber fields could be built 
in the Marianas. 

The recommendation to the Combined Chiefs noted that in response to the 
Air Plan for the Defeat of Japan CBI Theater had suggested basing eight B-29 
groups at Calcutta and staging them through Cheng-tu. The Combined Staff 
Planners had not weighed this proposal in detail but thought it might well be 
feasible. Their own plan called for sending 2,000 B-24's to India immediately 
after Germany's defeat and with them flying supplies to China to begin 
preparations for the reception of B-29's en masse. 12 

With the CCS advisers thinking of a major effort through the Pacific and 
of bypassing China, criticism of existing strategy for the mainland of Asia 
developed. As defined by the Strategy Section of the Operations Division 
(OPD), the current plan called for keeping China in the war as an effective 
ally in order to use Chinese bases to bomb the Japanese islands. A great con- 
verging attack from east and west was contemplated, to open the Hong Kong- 
Canton area as a base from which to launch a drive that would open a line of 
communications to the North China Plain. This strategy seemed defective 
because it was not co-ordinated with the major effort being planned for the 
Pacific, which included bombing Japan from the Mariana Islands in January 
1945 and launching the final air and amphibious assault on the Japanese home- 
land not later than mid-1946. The plan of securing Chinese bases seemed too 
costly in men and materiel for the advantages it would yield, mainly, the chance 
to bomb Japan. Using Chinese bases to the fullest extent would probably 
require the conquest of all Burma in order to reopen the line of communica- 
tions from Rangoon northward. The Strategy Section, OPD, considered that 
the situation in Asia, despite all earlier efforts, continued to be bad. China was 
still an ineffective ally, and Indian forces could not mount a major offensive. 
The Assam line of communications was still no better. Japan was improving 
her defensive position, while current U.S. strategy in Asia called for no effective 
blow at Japan proper before 1946. 

Therefore, the Strategy Section of OPD recommended that the present 
approved undertakings to keep China in the war as an effective ally be fulfilled; 
that a limited bomber offensive from China be mounted as insurance for the 
Pacific effort; that no further commitments be made to CBI Theater; and that 
no more than thirty Chinese divisions be trained and equipped, plus three more 
divisional sets of equipment to be used in beginning the training of the Second 

12 CPS 86/2, 25 Oct 43, sub: Defeat of Japan Within Twelve Months After Defeat of Germany. 



Thirty Divisions in east China. The report was innocent of diplomatic consid- 
erations; its thought was that the goal of the Pacific war was the military defeat 
of Japan. 13 

The next voices raised were those of the members of the Joint Strategic 
Survey Committee (JSSC), placing their views before the JCS on the eve of 
Sextant. A small group of distinguished senior officers, taking the broad 
detached view, they spoke with the weight of long experience. Though they 
were in general agreement with the Quadrant decisions, they did think these 
should be reappraised in the light of the recent studies of the problem of speed- 
ing Japan's defeat, which had shown the great importance of taking the 
Marianas as bases for the B-29's. The JSSC stated: 

We feel that without depreciating the importance of the effort against Japan by way of 
China, the key to the early defeat of Japan lies in all-out operations through the Central 
Pacific with supporting operations on the Northern and Southern flanks, using all forces, 
naval, air, and ground, that can be maintained and employed profitably in these areas. We 
believe that this principle and the related principle that operations from the West (via 
Singapore) would be of a diversionary nature have not been sufficiently recognized and 
emphasized. 14 

Therefore, by the time the Sextant Conference met, important agencies 
among the United States' planners were counseling a reappraisal of the United 
States strategy. Had the Chinese been zealous and industrious in preparations 
for a campaign in Burma, had they accepted and carried out Stilwell's sugges- 
tions for a potent Chinese Army of sixty divisions, and had the Generalissimo 
in March 1943, against whatever odds, crossed the Salween River into Burma, 
the United States would have been morally obligated to support the Chinese in 
projects it had persuaded them to undertake. Nor could India Command have 
held back if Chinese troops tried to liberate a major portion of the Common- 
wealth. But the Chinese had not thought in those terms, the months had gone 
past, and now American planners were beginning to conclude that they could 
defeat Japan without Chinese bases and without a rejuvenated Chinese Army. 
The recommendations which the Strategy Section of OPD made to arm thirty- 
three Chinese divisions, in November 1943, complemented the conclusions that 
Stilwell had reached one month before. Stilwell's superiors were quietly dis- 
carding the mission they had given him in February 1942, "to assist in improv- 
ing the combat efficiency of the Chinese Army"; by implication, other tasks 
would be forthcoming. 15 

The U.S. advance across the Central Pacific began 20 November when U.S. 
forces landed in the Gilberts group. After seventy-two hours of fighting, some 
of it of the most desperate nature, the Marines had their objective. American 
sea power had taken a giant stride closer toward Japan. 

13 Memo, Col Joseph J. Billo, Chief, Strategy Sec OPD, for Chief, Strategy and Policy Gp OPD, 
4 Nov 43, sub: Reanalysis of Our Strategic Position in Asia. ABC 337 (18 Oct 43) Sec 5, A48-224. 

14 JCS 533/5, 8 Nov 43, sub: Recomm e nded l. ine f Action at Next U.S. -British Stf Conf. 
n Stilwell's Mission to China, |2hs. ll| an j X. | 



The Chinese Prepare for SEXTANT 

Having expressed his opinions on the proposed plan for SEAC's share in 
Burma operations, Stilwell left his liaison personnel to participate in the final 
discussions and returned to Chungking to inform the Generalissimo of the 
trend of SEAC's thinking and to prepare with him for the forthcoming meeting 
between the President and the Generalissimo. The Generalissimo was markedly 
pleasant and co-operative. After the events of October Stilwell was extremely 
skeptical of the Generalissimo's sincerity, but work must be done before the 
forthcoming conference, and Stilwell applied himself to it. 16 

At the suggestion of his friends, Mesdames Chiang Kai-shek and H. H. 
Kung, Stilwell, as Joint Chief of Staff, China Theater, on 5 November 1943 
prepared and submitted a report to the Generalissimo on SEAC planning and 
Y-Force's progress in its preparations to attack from Yunnan. Telling the Gen- 
eralissimo that no final SEAC plans had been made, Stilwell pointed out that 
"it is certain" the Chinese would be expected to make a converging attack from 
Assam and Yunnan into north Burma. "If for any reason the Y-force does not 
attack, the British [military] will have an excellent argument for giving up any 
plans for reopening communications with China. They have contended that the 
Chinese army is incapable of fighting and that there is no use in trying to build 
it up; failure to fight now will tend to prove them right. ..." Then Stilwell 
explained why the Y- Force was not ready: 

3 (A) The long delay in furnishing replacements has left all divisions far below 
strength. . . . 

(B) The training has not yet reached the bulk of the men. . . . 

(C) The equipment brought in from India has not been distributed. There has been 
trouble in getting the Chinese supply agencies to take this equipment, and unusual delay in 
getting it into the hands of the troops. Some divisions are so weak that they cannot take care 
of their quota. 

(D) The majority of the men are physically incapable of sustaining prolonged hard- 
ship. . . . 

(E) The high-ranking officers generally have no offensive spirit. . . . 

(F) Insufficient trucks and animals have been provided. [Stilwell asked that the General- 
issimo issue the necessary corrective orders in the most forceful manner, and closed by 
warning that] It is too late already for half measures, or further delays; where a few months 
ago corrective measures could have been taken in an orderly manner, it is now too late for 
any but the most drastic and thorough -going action. 17 

The Generalissimo took this candor in good part. He promised 50,000 
replacements to bring the Y-Force up to strength, plus extra rations to meet 
the problem of malnutrition. The Chinese leader's cordiality was marked. 18 It 

16 (\) \Stilwelh Mission to China, Ch. x| (2) The Stilwell Papers, pp. 237-38. (3) In his Black 
Book, 6 November 1943, Stilwell wrote: "Is this real cooperation, or am I going goofy? . . . The 
catch is probably that he's willing but the blocking backs in the War Ministry will throw us for a 
loss. But just now, we are all honey and sweetness."' 

17 Memo, Stilwell for Generalissimo, 5 Nov 43. Stilwell Documents, Hoover Library. 

18 The Stilwell Papers, pp. 237-38. 



extended to StilwelPs suggestions for the Chinese proposals to be offered at 
Sextant. Possibly Stilwell hoped that if the Chinese leader offered such a pro- 
gram to the President and the Prime Minister, the Generalissimo himself 
would be obliged to adhere to it. And, faithful to the "bargaining" policy that 
he always wanted to follow, Stilwell spelled out what China should expect of 
her Allies if she did her part. 

MEMORANDUM: His Excellency, Generalissimo Chiang Kai-shek 

The Generalissimo's program is to bring up to effective strength, equip, and train 90 combat 
divisions, in 3 groups of thirty each, and 1 or 2 armored divisions. 

1. The first group consists of the divisions in India, and those assigned to the Y-force in 
Yunnan Province. These divisions should be at full strength by January 1, and by that date 
satisfactorily equipped. . . . 

2. The second group of thirty divisions has been designated [note that these are sug- 
gested proposals to be adopted by the Generalissimo, not a recital of accomplished facts] and 
a school has been set up. . . . With a road to India open, [the second thirty divisions] 
should be re-equipped and ready for the field in August of 1944. 

3. A similar process will be followed with the third group of 30 divisions with target date 
of January 1, 1945. After the reopening of communications through Burma, 1 or 2 armored 
divisions will be organized. 

4. All resources available in China will be used to produce effective combat units. 
Trained men of existing units will be made available as fillers. 

5. China will participate according to the agreed plan in the recapture of Burma by 
attacks from Ledo with the X-force [Ledo force] and from Paoshan with the Yunnan force. 
This operation will be supported by naval action in the Bay of Bengal. Before the operation, 
British naval forces should be concentrated in time and fully prepared for action. 

6. The training program will be followed and intensified. 

7. Necessary airfields will be built and maintained. 

8. In the event that communications are reopened through Burma and necessary equip- 
ment is supplied, an operation will be conducted to seize the Canton-Hongkong area and 
open communication by sea. 

The Generalissimo expects that: 

1. Before the 1944 rainy season an all-out effort will be made by the Allies to re-open 
communications through Burma to China, using land, air, and naval forces. 

2. The U.S.A. will supply the equipment for the three groups of 30 divisions, and the 
armored divisions. 

3. The Fourteenth U.S. Air Force will be maintained as agreed and supplied sufficiently 
to allow of sustained operations. 

4. The Chinese Air Force will be built up promptly to 2 groups of fighters, 1 group of 
medium bombers, 1 reconnaissance squadron, and 1 transport squadron, and maintained at 
that strength. By August of 1944 a third group of fighters, and a group of heavy bombard- 
ment will be added and maintained thereafter. 

5. Following the seizure of the Canton-Hongkong area, the U.S. will put 10 infantry 
divisions, 3 armored divisions and appropriate auxiliary units into South China for opera- 
tions against Central and North China. Contingent upon this allocation of troops, the 
Generalissimo will appoint American command of those units of the combined U.S. Chinese 
[sic] forces which are designated in the order of battle, under his general direction. 



6. The U.S. will, at the earliest practicable time, put long-range bombing units in China 
to operate against the Japanese mainland. 

7. The ferry route will be maintained at a capacity of at least 10,000 tons a month. 

8. Training personnel will be supplied as required. 

9. Medical personnel will be supplied for the second and third groups of divisions. 
For the Generalissimo, 

Joseph W. Stilwell, 
Joint Chief of Staff for Generalissimo. 19 

Stilwell thus proposed that the Generalissimo ask the United States to train 
and equip no less than ninety Chinese divisions. So imposing a force would 
dominate Asia south of the Amur River. Only the Red Army in Siberia could 
have faced it, and even then, the issue would have been uncertain. The General- 
issimo was apparently favorably impressed by Stilwell's suggestions, for many 
of them were offered on behalf of China at the Sextant Conference. 20 

Confirming the Generalissimo's cordiality, Madame Chiang telephoned 
Stilwell that night. She told the American general that the Generalissimo was 
"not only pleased but happy," over his conference with Stilwell. 21 On 
7 November Stilwell saw the Chinese Chief of Staff, Gen. Ho Ying-chin, who 
was not encouraging about replacements, but presumably General Ho had not 
yet received orders from the Generalissimo. 22 

Four days later, on 11 November, General Stilwell, General Hearn, and Col. 
John E. McCammon, G-3, Chungking, met with General Ho and two of his 
staff at the Chinese National Military Council to receive the Generalissimo's 
formal answer to Stilwell's 5 November memorandum. The National Military 
Council agreed to a converging attack on Burma by British and Chinese troops 
but desired to hold their own advance until the British were actually attacking 
Kalewa in Burma. On replacements, the Chinese said that 35,000 were en route 
to Yunnan. In addition, 54,000 more men would be sent. To move them, the 
Chinese would need motor fuel, which Stilwell promptly undertook to furnish. 
The Chinese agreed to provide more food for the Y-Force. Their medical needs 
were presented. The questions of interpreters, spare parts, artillery horses, and 
7.92-mm. ammunition were all presented affirmatively and solutions speedily 
agreed on by both sides. 23 Simultaneously with these conferences on military 
matters Stilwell found time to talk with General Hurley, now in Chungking on 
behalf of the President to arrange for the Generalissimo's visit to Cairo. Gen- 
eral Hurley made an excellent impression on Stilwell, who enjoyed Hurley's 
anecdotes and his comments on Allied powers and personages. For his part, 

19 Stilwell Documents, Hoover Library. 

20 That the Generalissimo returned a written answer is implied in The Stilwell Papers, p. 240. 
However, the authors have not been able to find it. 

21 The Stilwell Papers, p. 236. 

22 Ibid., p. 238. 

2J Notes, Conf, NMC, 11 Nov 43. Marginal notes show action directed for Americans. Stilwell 
Documents, Hoover Library. 



Hurley liked the outspoken, acidly witty Stilwell, and the two men got on very 
well. 24 In speaking to the Generalissimo, Hurley gave a brief review of U.S. 
policy, which included "belief in a "free, strong, democratic China, predominant 
in Asia." 25 

Thus, on the eve of Sextant the opportunity of creating an effective Sino- 
American effort in Asia seemed to exist. In October StilwelPs diaries showed 
the utmost skepticism about the Generalissimo's desire to reform his Army and 
use it aggressively against the Japanese. But now the Generalissimo was again 
receiving Stilwell's views, he was considering them favorably, and he was over- 
ruling his subordinates and ordering them to take action, a changed attitude 
which can be seen in the great difference of General Ho's expressions of 7 No- 
vember from those of 1 1 November. For his part, as the marginal notes on the 
1 1 November minutes show, Stilwell was meeting every Chinese proposal and 
promise with appropriate orders to his own people. If this atmosphere per- 
sisted, Stilwell and the War Department might be moved to re-examine their 
conclusions of October and November 1943. 

The issue of Sino- American relations was about to move out of Stilwell's 
hands into those of his superiors, the President, the Prime Minister, and the 
Generalissimo. At Sextant it would be up to the United States and the British 
Commonwealth to abide by the pledge of Trident that nothing would be left 
undone to relieve the siege of China. If the President and the Prime Minister 
made good on the plans for a major Allied operation in Burma, Sino- American 
co-operation could flourish. If, however, the Generalissimo was given reason 
to be dissatisfied with what he received from the President and the Prime Min- 
ister, then Stilwell's position would be compromised. If the bases of Sino- 
American co-operation were not present, Stilwell's personal efforts could do 
little to remedy the situation. 

Presenting CHAMPION at Cairo 

With General Hurley in Chungking, the myriad details attendant on the 
flight to Cairo of the Generalissimo, Madame Chiang, and their entourage 
were speedily worked out. It was agreed among the powers that Mr. Roosevelt 
and Mr. Churchill would meet the Chinese leader in late November and then 
confer with Marshal Stalin in Tehran, Iran. The Combined Chiefs of Staff 
would be present and so would Admiral Mountbatten and Generals Stilwell, 
Chennault, and Wedemeyer. 

Stilwell arrived at Cairo on 20 November. The following day he was able 
to see General Marshall in company with General Hurley and General Somer- 

24 The Stilwell diaries of this period have several appreciative comments on General Hurley. 
Hurley's recollections of his first meetings with General Stilwell were given to the authors. Intervs 
with Gen Hurley, Jan 49, Feb 50. 

25 The Stilwell Papers, pp. 238-40. 



CAIRO CONFERENCE participants were, left to right front, Generalissimo Chiang 
Kai-shek, President Roosevelt, Prime Minister Churchill, and Madame Chiang, and, 
standing left to right, Gen. Shang Chen, Lt. Gen. Lin Wei, General Somervell, General 
Stilwell, General Arnold, Field Marshall Sir John Dill, Admiral Mountbatten, and 
Maj. Gen. Adrian Carton de Wiart. 

veil of Army Service Forces. Stilwell was anxious to raise many points with 
Marshall, presumably before the conferences began. His notebook records 

Minister] of War. (replace [Gen Ho Ying-chin]). U.S. command after Pacific port 
[is opened]. 90 divisions. Offensive-defensive alliance. SEAC ambitions [to absorb CBI The- 
ater]. Mountbatten wants me out. U.S. command of U.S. units. After Champion? Future 
[of] CBI. 

Louis [Mountbatten]: (1) Wants authority over ATC so as to "protect" it; (2) Wants 
China air plans for '44 and '45; (3) Wants responsibility for operation of Burma Road; (4) 
Liaison with Miles [U.S. Naval Observer Group in China]; (5) De Wiart [British liaison 
to Generalissimo] in our hqs; (6) Liaison offs [officers] with Chinese [divisions]; (7) 
Wants to absorb Rear Echelon; (8) Squadron of Spitfires to China; (9) Air staff" mission; 
(10) Medical mission. 

Claims GCM [Marshall] and Arnold told him to integrate [the Anglo-American air 
forces in India]. 

The plan for Champion: Piece meal; indefinite objective; Indaw abortion. No problem. 



Utopia [seizure of Andaman Islands] abortion, no bearing; leaves Chinese to hold sack; 
no British troops— unreliable Indian troops. 26 

Whether Stilwell presented these points at one session, or how Marshall 
reacted to them, is unknown. In his talk with Marshall, Hurley, and Somer- 
vell, Stilwell was warned that the President highly disapproved of his disre- 
spectful references to the Generalissimo. 2 " 

The first plenary session of Sextant was set for 1100, 23 November 1943. 
The Joint Chiefs of Staff met briefly with Stilwell and Wedemeyer before the 
plenary session to receive their comments on Champion, SEAC's plan for 
Burma. No attempt was made to weigh the plan of Champion, which had 
been adopted over Stilwell's objections. Of the airborne operation, he remarked 
that he saw no point in cutting Hump tonnage just to drop a division in the 
jungle during the rains. Stilwell did not think the Japanese line of communica- 
tions to Myitkyina a vital one and did not want it blocked at the expense of 
Hump tonnage (which would embarrass his relations with the Generalissimo 
and Chennault). However, Stilwell pledged that once Champion began, he 
would do his best to carry it out. Wedemeyer commented that while Cham- 
pion did provide attacks on all key points, he did not particularly care for the 
Arakan situation, in which two divisions plus two brigades were given only 
the most limited objectives, for he mistakenly believed they faced but two Jap- 
anese regiments. Actually, the Japanese 54th Division was then moving up to 
join the 55th in the Arakan. 

Stilwell's comments prefaced his presentation to the Joint Chiefs of Staff 
of the Generalissimo's and his proposals for China Theater, based on Stilwell's 
paper, Proposals for the Coming Conference. The Generalissimo called for oc- 
cupation of north Burma, intensive training of the Chinese Army, and im- 
provement of the line of communications to China. He desired B-29 opera- 
tions from China Theater in early 1944, air attacks in the Formosa-Luzon area 
in October 1944 to support U.S. naval operations in that area, the taking of 
Canton and Hong Kong in November 1944-May 1945, and an attack on For- 
mosa from Chinese ports, if required. The paper was most significant because 
it had the Generalissimo's approval. This was, so Marshall said, the first time 
since the war began that the Generalissimo had shown an active interest in the 
improvement and employment of his Army. General Marshall and Admiral 
Ernest J. King, U.S. Chief of Naval Operations, thought this attitude extremely 
important and not to be discouraged if at all possible. 28 After this session 

26 (1) These jottings are from one of Stilwell's notebooks of the type in which he kept his 
diary. This one is labeled Data, and is hereafter cited as Data Notebook. At the top of the page 
on which these entries begin, Stilwell wrote "GCM" in bold letters. (2) Stilwell's fears about 
Mountbatten's attempts to whittle away his authority in India and China are also expressed in Rad 
AGWAR863, Stilwell to Marshall, 11 Nov 43. Item 1234, Bk 4, JWS Personal File. 

27 The Stilwell Papers, p. 245. 

28 (1) See pp. 57-58, above. (2) Min, JCS 128th Mtg, 23 Nov 43, Item 2. (3) CCS 405, 
22 Nov 43, sub: Role of China in Defeat of Japan. (4) Japanese Study 89. 



closed, the American service chiefs joined their colleagues and superiors for the 
plenary session. 

Admiral Mountbatten had expected Champion to be first presented to the 
British Chiefs of Staff by himself, and to the Joint Chiefs of Staff by General 
Wedemeyer. On their approving it, Champion would go to the Combined 
Chiefs of Staff, and if they concurred, be presented to the Generalissimo, the 
President, and the Prime Minister as an agreed-on CCS proposal. This was the 
usual practice in such cases, but at Sextant it was reversed. The Generalis- 
simo was present, though unfortunately for security reasons his arrival was not 
announced in advance, so neither the President nor the Prime Minister had 
been at the airport to greet him and Madame Chiang. This was a blow 
to Chinese pride. 

Because the Generalissimo was at hand, and because Roosevelt and Church- 
ill wanted him to enter immediately into military discussions, the SEAC plan 
was laid before the Generalissimo at once, and therefore without its having 
been considered by the CCS. Thus, the Generalissimo was being asked to ap- 
prove Champion in advance of its approval by the Allies. 

As presented formally to the three Allied statesmen, to Harry L. Hopkins, 
Madame Chiang, and the highest service advisers, Champion's first phase 
called for the advance of the Chinese 22d and 38th Divisions from Ledo, an 
operation then under way. In mid-January 1944, 15 Corps would move for- 
ward in the Arakan to take up an improved line, and would exploit any suc- 
cess that might be gained. At the same time 4 Corps would advance on Maw- 
laik, Minthami, and Sittaung, driving southeast as far as possible. In February 
1944 three long-range penetration groups would attack. Paratroops would seize 
Indaw in mid-March after which the 26th Indian Division would fly in to hold 
it. 29 A major amphibious operation would be staged in the Bay of Bengal. For 
security reasons, the amphibious operation was not further described to the 
Chinese. As for weather, Mountbatten hoped to end his advance by early April 
when the monsoon rains would break. During the monsoon, the long-range 
penetration groups would operate, and if the CCS gave the needed resources, 
the advance would resume after the monsoon's end. The rains were expected 
to prevent a Japanese reaction. 30 

The Chinese, apprised of Champion weeks before by Stilwell, were imme- 
diately critical. The Generalissimo did not believe that 15 and 4 Corps were 
intended to advance far enough into Burma; he wanted them to drive on Man- 
dalay. He insisted that the advance must be synchronized with a naval oper- 
ation. But the Generalissimo's argument for a naval operation was now affected 
by a sovereign fact which he disregarded. The Japanese were known by the 
Sextant conferees to have completed a railway from Thailand to Burma which 

29 ( 1) SEAC War Diary, 23 Nov 43. (2) Min, Sextant Conf, First Plenary Mtg, Villa Kirk, 
23 Nov 43. (3) Henry H. Arnold, Global Mission (New York: Harper & Brothers, 1949), p. 460. 

30 ( 1) Min, CCS 129th Mtg, 24 Nov 43, Item 5. (2) CCS Info Memo 166, 18 Dec 43. 



made them independent of imports through Rangoon. The Generalissimo also 
insisted that whatever the demands of Burma operations the Hump lift must 
not fall below 10,000 tons a month. A day later Chennault gave the monthly 
requirements of the Chinese Air Force and the Fourteenth Air Force, 10,000 
tons a month. Asked by General Arnold what that would leave for the Chinese 
Army in China, which had a major role to play in the reconquest of Burma, 
General Chennault simply replied that 10,000 tons was what he needed. 31 

Trying To Reach Agreement 

These viewpoints having all been expressed, the conferees had two delicate 
tasks to handle simultaneously: to settle on a plan and to secure the Generalis- 
simo's assent to it. Reversing the usual process by which plans were approved, 
in order to spare the Generalissimo's feelings, was leading into ever more 
tangled thickets. Mountbatten was sent to the Generalissimo's villa to explain 
that if the offensive toward Mandalay which the Chinese leader desired was 
carried out, it would entail diversion of all Hump tonnage. "Welcome change 
from telling me to fix it up," wrote Stilwell. 32 

As Admiral Mountbatten tried to explain the situation, the Generalissimo 
grew enthusiastic and announced he would press for both an airborne assault 
on Mandalay and 10,000 tons a month over the Hump, which would require 
an added 535 transports sent to India. Mountbatten finally escaped by saying 
that he would lay the Generalissimo's wishes before the CCS to see if they 
could find the 535 transports, which Mountbatten knew were nowhere to be 
had. The CCS formally stated that the 535 aircraft could not be found, and in 
view of the uncertainty surrounding the Generalissimo's attitude, Mountbat- 
ten was asked to obtain his formal agreement to go back into Burma. 3 ' 

While Mountbatten, aided by Churchill, was essaying this task, Stilwell 
went with Marshall on 25 November to confer with the President. Before the 
interview Stilwell noted what he wanted to say to the President about the prob- 
lems that faced him in China: "Ask FDR: Field chief of staff [to Generalis- 
simo], can [have]: (1) Man power; (2) Executive authority; (3) U.S. troops; 
(4) Chinese- American command. Keep X-force, add one corps [as a force di- 
rectly under Stilwell's command]" 34 

Preparing for his interview with the President, Stilwell sketched a point he 
wanted to make: 

" (1) Min cited n. 29(2). (2) Min, CCS 129th Mtg, 24 Nov 43, Item 7. 

« (1) The Stilwell Papers, p. 246. (2) SEAC War Diary, 24 Nov 43. 

" (1) SEAC War Diary, 24 Nov 43. (2) Min, CCS 130th Mtg, 25 Nov 43, Item 1. 

34 Data Notebook. A little earlier in the Data Notebook, in the first version of his notes for the 
conference with the President, Stilwell put it: "FDR. Recommendations. Private army of one corps. 
Keep X-Force and add one in China. Recommend to Peanut more power for me. Field Chief of 
Staff. Oust running dog. FDR. My mission complicated by not knowing what direct messages 
[from Roosevelt to Chiang] contain No bargaining power. (Twilight)." 



No matter what Peanut agrees to, if something is not done about the Chinese high com- 
mand the effort is wasted. I suggest stipulation of U.S. command, with real executive au- 
thority. If impossible over large group, then over composite Chinese- American corps. Lack 
of real power and control of Gmo. He will order. Kan pu will block. Suggest new Minister 
of War or thorough re-organization of [Chinese] War Department. Or American take over the 
first 30 complete and operate them [Stilwell's italics]. 35 

Stilwell and Marshall entered the President's room, and Stilwell began his 
presentation. The President seemed to hear him with "little attention" and in 
the middle of Stilwell's report broke in to talk about the Andaman Islands, on 
which he wanted to put some heavy bombers. Trying to bring the discussion 
back to China's problems, Stilwell pled for some U.S. combat troops in CBI. 
In reply, the President offered to put a brigade of U.S. Marines in Chungking. 
"Marines are well known," said the President. "They've been all over China, 
to Peking and Shanghai and everywhere. The Army has only been in Tientsin." 

Stilwell told the President that the Chinese had reneged on their agree- 
ments, that to carry out his mission he needed more power and executive au- 
thority over Chinese troops. Stilwell also dwelt on the "basic factors of our 
presence" in China, that is, the Chinese were to supply the men while the 
Americans supplied weapons and training. The President, though promising to 
speak to the Generalissimo at once on these points, seemed to show little 
interest. 36 

The President's attitude depressed Stilwell, but the conference was not all 
negative. Mr. Roosevelt stated that the Generalissimo had agreed to Cham- 
pion. 37 Then came bigger news. An American corps was out of the question, 
but the Chinese could have equipment for ninety divisions and could help oc- 
cupy Japan. At the JCS meeting that day General Marshall had remarked that 
there was pressure on the President to give the Generalissimo something to 
show as a result of his trip, that the President had been spoken to about arm- 
ing the third thirty divisions but had postponed any definite commitment, 
though Roosevelt had made it clear the United States intended eventually to 
equip ninety Chinese divisions. 38 Now the President told the Generalissimo's 
joint chief of staff of the ninety-division intention, and Stilwell duly listed it 
among the "Cairo results." 39 

Returning to his quarters, Stilwell took the notes he had prepared for his 

35 Data Notebook. 

36 (1) Handwritten pages headed Story of J. Peene, Sr. (Hereafter, Story of J. Peene, Sr.) Stil- 
well Documents, Hoover Library. (2) Joseph Peene was General Stilwell's maternal grandfather. 
Mr. Peene was famous in family tradition for paying his employees in gold pieces. Because the Gen- 
eralissimo later asked the President for $1,000,000,000 gold the event may have reminded the 
sometimes waggish Stilwell of this episode from the days of the gold standard. See letter, Mrs. 
Winifred A. Stilwell to Sunderland, 4 August 1952. OCMH. (3) The Stilwell Papers, p. 246. 

37 The Stilwell Papers, p. 246. 

38 (1) Story of J. Peene, Sr. The exact wording is: "What shall we give the Chinese? Equip, 
for 90 XX [divisions]. But the American Corps is out, and we give them Japan. What a laugh 
for the Japs." (2) Min, JCS 130th Mtg, 25 Nov 43, Item 6. 

39 Data Notebook. 



talk with the President, drew a line diagonally across the page and wrote above 
them: "NB: FDR is not interested." 40 

While Stilwell was preparing to meet with the President, Mountbatten and 
the Prime Minister attempted to secure a firm assent from the Generalissimo to 
Champion. Initially, as the President told Stilwell, they succeeded. On the 
early afternoon of 25 November the Generalissimo agreed to go into Burma 
on two conditions: that the Royal Navy's Eastern Fleet command the Bay of 
Bengal, and that an amphibious operation be mounted there. That evening 
the Generalissimo met again with the President and reversed himself on every 

Mountbatten was again sent into action to restore the situation but found 
the Generalissimo obdurate. So Mountbatten turned to Churchill, had lunch 
with him, and the Prime Minister agreed that he with the President and 
Madame Chiang would try to bring the Generalissimo round. The Allied lead- 
ers met the afternoon of the 26th at tea, unfortunately with neither secretaries 
nor minutes. After tea the Prime Minister and Madame Chiang separately told 
Mountbatten that the Generalissimo had agreed on every point. Such was the 
situation when Churchill and Roosevelt with their key advisers departed for 
Tehran, and the Generalissimo prepared to go to Chungking. For the first time 
in the war, the Prime Minister, the President, and the CCS had met the Gen- 
eralissimo and endeavored to secure a binding agreement from him. "They 
have been driven absolutely mad," wrote Admiral Mountbatten, "and I shall 
certainly get far more sympathy from the former in the future." 41 

With the dignitaries out of the way, Admiral Mountbatten called a meet- 
ing of the SEAC delegation on 27 November to clear up the loose ends. He 
felt "staggered" when Stilwell came in to tell him that just before departing 
that morning the Generalissimo had reversed himself again, rejected all his 
previous agreements, and ordered Stilwell as the latter put it, to "stay and pro- 
test. I am to stick out for Toreador [the airborne assault on Mandalay] and 
10,000 tons [a month over the Hump]." 42 Mountbatten thought quickly. He 
had arranged to inspect the Ramgarh Training Center together with the Gen- 
eralissimo in a few days and believed that if he had the elusive Chinese leader 
to himself for a few minutes he might succeed in getting a binding agreement 
from him. So he became diplomatically deaf, told Stilwell he had not under- 
stood him, and asked that a radio be sent to him at New Delhi. 43 

Summing up the Sextant Conference at that point, Stilwell asked him- 
self: "So where are we? Tarzan? Tonnage? Command? Sure on equipment for 
90 divisions. . . ." 44 

40 Ibid. 

41 (1) Extracts, SAC's Personal Diary, 25, 26, 27 Nov 43; quotation from Extract, 27 Nov 43. 
SEAC War Diary. (2) The Stilwell Papers, p. 246. 

42 (1) SEAC War Diary, 27 Nov 43. (2) Stilwell Diary, 27 Nov 43. 

43 SEAC War Diary, 27 Nov 43. 

44 Stilwell Diary, 27 Nov 43. 



Thus, of the two delicate and simultaneous operations, the agreement and 
the plan, one had not been brought off. Nor was there agreement between the 
President, the Prime Minister, the CCS, and the JCS on future operations in 
SEAC. Churchill early indicated his attitude by telling Admiral Mountbatten 
on 21 November that he meant to have a landing on Sumatra or nothing, that 
if there was no such operation, he would take away SEAC's landing craft for 
an operation against the island of Rhodes in the Mediterranean. 45 A few days 
later Marshall remarked that Roosevelt had expressed his opposition to any 
diversion of Royal Navy landing craft from Buccaneer (now the code name 
for the Andamans operation which was to meet the Generalissimo's long- 
standing demand for an amphibious operation). This expression also was the 
view of the Joint Chiefs, who were strong for Buccaneer. In a conference at 
Tehran between the President and the JCS, it was observed that the British 
would do all they could to cancel Buccaneer for an operation against Rhodes. 
The President quickly replied that the Allies were obligated to the Chinese to 
stage Buccaneer, an attitude which suggests that he was unaware of the Gen- 
eralissimo's final reversal. However, at the first CCS session at Tehran the 
British Chiefs of Staff urged the abandonment of Buccaneer, and it remained 
to be seen whose view would prevail. 46 

While the President was at Tehran, the Cairo Declaration was issued by the 
President and the Prime Minister as a joint pronouncement of the United 
States, the British Commonwealth, and China. In sharp contrast to the actual 
course of events at Sextant, the declaration read: "The several military mis- 
sions have agreed upon future military operations against Japan. The Three 
Great Allies expressed their resolve to bring unrelenting pressure against their 
brutal enemies by land, and sea. This pressure is already rising." The declara- 
tion went on to pledge the return of Manchuria, Formosa, and the Pescadores 
to China, and that Korea should be free and independent. It then concluded: 
"With these objects in view, the three Allies, in harmony with those of the 
United Nations at war with Japan, will continue to persevere in the serious 
and prolonged operations necessary to procure the unconditional surrender of 
Japan." 47 

While the President and Prime Minister were meeting with Marshal Stalin 
at Tehran, the Generalissimo again changed his mind about Burma operations. 
While inspecting the Chinese New First Army at Ramgarh on 30 November, 
he again agreed to join in Champion. He confirmed his resolve in a speech to 
the Chinese soldiers, placing them under Mountbatten and Stilwell for the 
coming operations. 

45 SEAC War Diary, 21 Nov 43. 

46 ( 1 ) Min, JCS 1 31st Mtg, 26 Nov 43, Item 3.(2) Min, Mtg of President and JCS, American 
Legation, 28 Nov 43. (3) Min, CCS 131st Mtg, 26 Nov 43, Item 4. (4) Min, CCS 132d Mtg, 30 
Nov 43. 

47 U.S. Department of State, United States Relations With China (Washington, 1949) p. 519- 



CHIANG KAI-SHEK AT RAMGARH. Accompanied by Madame Chiang and 
Admiral Mountbatten the Generalissimo inspects the Chinese New First Army. 

I feel greatly inspired today as I am here with you, officers and men, at this post. Being 
able to speak to you in a friendly land, is indeed, a rare opportunity. You must pay full at- 
tention to every word I say and bear it firmly in your mind. It shall serve as a moral en- 
couragement for your endeavor to glorify our nation by adding a glorious page to the his- 
tory of our national army. Now that our National Army is enabled to come over to India as 
a combined combat strength with our worthy allies, [it?] has already registered an illus- 
trious page in our national annals. 

It is also your good fortune that you are placed under the joint command of Admiral 
Lord Louis Mountbatten and General Joseph W. Stilwell, respectively supreme commander 
and deputy supreme commander of S.E. Asia Command. My expectation of the New First 
Army is for you to accomplish this worthy mission. My meeting with you here today is just 
like a family reunion which imparts profound attachment to both father and sons. It is there- 
fore your duty to listen to my words as follows [here, the Generalissimo encouraged his 
troops to fight well for China]. I exhort you to keep my words. Unitedly under the joint 
command of Admiral Lord Mountbatten and General Stilwell you shall destroy the 
enemy. . . , 48 

Over the Watershed: The Changed Attitude Toward China 
At Tehran the President met Marshal Stalin for the first time. Explaining 
his China strategy, the President spoke of converging attacks on north Burma, 

48 Address. Generalissimo to Chinese New First Army in India, 30 Nov 43. JWS Misc Papers, 
1943. (See Bibliographical Note.) 



and of amphibious operations in the Bay of Bengal. The goal was to open the 
road to China and supply China so that it would stay in the war and, also, to 
put the Allies in a position to bomb Japan from Chinese bases. Marshal Stalin 
expressed no opposition to this, and, indeed, repeated his earlier promises to 
enter the war against Japan. 49 

After meeting and conferring with Marshal Stalin, the President, in the 
opinion of Robert E. Sherwood, arrived at certain conclusions with regard to 
the Soviet Union and its leader: 

Roosevelt now felt sure that, to use his own term, Stalin was "getatable," despite his 
bludgeoning tactics and his attitude of cynicism toward such matters as the rights of small 
nations, and that when Russia could be convinced that her legitimate claims and require- 
ments—such as the right to access to warm water ports— were to be given full recognition, 
she would prove tractable and co-operative in maintaining the peace of the postwar world. 

If, therefore, good relations could be established with the Soviet Union, all the 
pieces of the postwar puzzle would fall into place. In the immediate present 
there was no doubt of what the Soviet Union wanted— a cross-Channel assault 
(Overlord) and a landing on the coast of southern France (Anvil) as soon 
as possible and on as big a scale as possible. 50 The President, therefore, would 
weigh operations in Southeast Asia in an atmosphere very different from that 
of the first conferences in Cairo a few days before. Such was the situation when 
the President, the Prime Minister, and the Combined Chiefs of Staff finished 
at Tehran and returned to Cairo. 

Mr. Churchill and the British Chiefs of Staff immediately attacked Buc- 
caneer. Churchill took the Stalin promise to enter the war with Japan as a 
stunning surprise that changed the whole strategic picture. He called it a de- 
cisive event. Soviet entry in the Pacific war would give the Allies better bases 
than China ever could. In the light of Stalin's promise, operations in South- 
east Asia lost much of their value. In this connection, he was astounded by 
SEAC's requirements for Buccaneer, which he understood to be 58,000 men 
to oppose 5,000 Japanese. The other decisive event, said Churchill, was setting 
the date for Overlord. Nothing anywhere should interfere with that great 
operation. The proper course, the Prime Minister argued, was to cancel Buc- 
caneer and use its landing craft to reinforce the amphibious assault on south- 
ern France, Anvil. 51 

The Prime Minister's pleased surprise at Marshal Stalin's promise to enter 
the Pacific war and his argument that because of it the strategic picture in the 
Pacific had changed since the first Cairo meetings were difficult to reconcile 
with the circumstance that the Soviet Union originally promised to enter the 

49 (1) Robert E. Sherwood, Roosevelt and Hopkins: An Intimate History (New York: Harper & 
Brothers, 1948) pp. 778-79. (2) CM-IN 1946, Ambassador W. Averell Harriman to Marshall, 
4 Nov 43. 

50 Sherwood, Roosevelt and Hopkins, pp. 788, 798-99. 

51 Min, Sextant Conf, Third Plenary Mtg, Villa Kirk, 4 Dec 43. 



Pacific war in October 1943 at the Moscow Conference and repeated its prom- 
ises in November. 52 

At some point during these post-Tehran discussions of Buccaneer, a radio 
from General Boatner in north Burma to theater headquarters in New Delhi, 
detailing at length the command problems he had met with the Chinese, ar- 
rived at Cairo. By mischance, it had been so forwarded, and was then delivered 
to the SEAC delegation. Circulated as an admission by Stilwell's own head- 
quarters that even U.S.-trained Chinese troops were unreliable the radio was a 
telling argument against any campaign that depended on the Chinese in any 
capacity. 53 

Mr. Roosevelt with Admiral King and Admiral William D. Leahy, the 
President's Chief of Staff, held that there was a definite commitment to the 
Generalissimo, and that a whole train of unhappy consequences might follow 
if China's allies broke their promise. He had a moral obligation to the Chinese, 
Roosevelt remarked, and could not forego the operation without a great and 
readily apparent reason. There the 4 December session ended, with a directive 
from the President and Prime Minister to the Combined Chiefs to try to find 
agreement on that basis. 54 The JCS met at 0900 on 5 December and found 
themselves still in accord on the need to execute Buccaneer. 

The Combined Chiefs met at 1030. General Marshall drew attention to a new 
strategic factor which had arisen since Trident. The blast of world-wide 
publicity following SEAC's creation had attracted heavy Japanese reinforce- 
ments to Burma which would seize the initiative unless the Allies struck first. 
Marshall feared that such a Japanese offensive would imperil the Hump route. 
If it would be possible to abandon Buccaneer and still carry out the North 
Burma Campaign, Marshall would not be seriously disturbed, but he did not 
think there would be a Burma campaign unless there was an amphibious opera- 
tion. Admiral Leahy remarked briefly that canceling the amphibious operation 
meant either the failure or the abandonment of the Burma campaign. 

The Chief of the Imperial General Staff, General Brooke, and Air Chief 
Marshal Sir Charles Portal repeated the arguments that Buccaneer was a 
diversion from the main effort in Europe and that the Chinese contribution was 
a negligible factor. They also noted that the main effort against Japan was now 
to be made in the Pacific, which was inconsistent with a heavy allocation of 
resources to Burma. The meeting ended with a decision to present the various 
points in dispute to the President and the Prime Minister. 55 

Mr. Roosevelt opened the plenary session by pointing out that Buccaneer 
was the dividing issue between the staffs. He acknowledged that the Gen- 

52 The original statement was made by Soviet Foreign Minister Vyacheslav M. Molotov to Cor- 
dell Hull at Moscow and confirmed shortly after by Molotov to Harriman, who promptly relayed 
it to General Marshall. (1) Cordell Hull, The Memoirs of Cordell Hull (New York: The Macmillan 
Company, 1948) II, 1309. (2) Rad cited n. 49(2). 

" (1) Story of J. Peene, Sr. (2) Ltr, Hill to Ward, 2 Sep 52. OCMH. 

54 (1) Min cited n. 51. (2) Min, CCS 134th Mtg, 4 Dec 43, Item 4. 

55 Min, CCS 135th Mtg, 5 Dec 43. 



eralissimo had left Cairo believing an amphibious operation would be carried 
out with Tarzan, the India-based portion of Champion. The President was 
dubious about staking everything on Russian good will, for he feared that the 
Allies might sacrifice the esteem of the Chinese without later securing the aid 
of the Russians. Admiral King rebutted the argument that Buccaneer had to 
be canceled to secure landing craft for Anvil by stating that a two-division lift 
for Anvil was in sight and might even be improved upon. This, he went on, 
would entail keeping back four months' production from the Pacific. 

Though the intimate connection between Buccaneer and Chinese partici- 
pation in Burma operations was admitted by all, it was quite clear that many 
of those present hoped the Generalissimo would perform his share of the bargain 
even though his Allies reneged on theirs. The British were adamant in oppos- 
ing Buccaneer as a diversion from Overlord, and Churchill made it clear 
that he felt no obligations to the Chinese. The meeting ended with an agree- 
ment to inquire of SEAC what it could do if the bulk of its landing craft were 
taken away. 56 

So questioned, SEAC quickly replied that canceling Buccaneer would, in 
the light of the Generalissimo's known attitude, lead to the collapse of Tarzan. 
In its stead SEAC suggested overland operations from Imphal toward Kalewa 
and Kalemyo in Burma (which if successful would be a long step toward 
Mandalay), continuation of the advance from Ledo, continuation of the current 
operations in the Arakan, and an assault by the long-range penetration groups 
at the proper time. SEAC acknowledged that this operation would not open the 
land route to China. 57 Admiral Leahy described SEAC's estimate of 50,000 men 
for Buccaneer as excessive, but General Wedemeyer replied that a smashing 
victory was needed to restore the morale of SEAC's troops and added that all 
the resources needed for Buccaneer, except an added 120 carrier-based fighters, 
were in sight. Admiral King immediately said that he might find four or six 
escort carriers to fill the gap. But there was still no agreement on Buccaneer, 
and the case went back to the President and the Prime Minister. 58 

On the night of 5 December Mr. Roosevelt accepted Mr. Churchill's argu- 
ments and withdrew his support from Buccaneer. In abandoning Buccaneer, 
the President overrode the very strongest protests of his service advisers. In his 
memoirs, Admiral Leahy wrote: 

I felt that we were taking a grave risk. Chiang might drop out of the war. He never had 
indicated much faith in British intentions, but had relied on the United States. If the Chinese 
quit, the tasks of MacArthur and Nimitz in the Pacific, already difficult, would be much 
harder. Japanese man power in great numbers would be released to oppose our advance 
toward the mainland of Japan. Fortunately for us, the courageous Chinese stayed in the fight. 

56 Min, Sextant Conf, Fifth Plenary Mtg, Villa Kirk, 5 Dec 43. 

57 Rad SEACOS 38, 6 Dec 43. Min, Sextant Conf, p. 312. 

58 Min, CCS 136th Mtg, 5 Dec 43, Item 1. 



After the war, in writing his memoirs, Admiral King remarked that he had 
been "distressed" by the breach of the long-standing promise to the Chinese, 
and added that in his opinion this was the only time during the war when the 
President had overruled the Joint Chiefs. 59 

After agreeing to cancel Buccaneer, the President and Hopkins drafted a 
radio to the Generalissimo telling him the bad news. The message was based 
on SEAC's estimate that there could be no major amphibious operation if 
Buccaneer was canceled. The estimate was in error, as SEAC soon discovered, 
but the two U.S. leaders naturally accepted it, and, consulting Churchill but 
not the CCS, told the Generalissimo there could be no successful amphibious 
operation simultaneously with Tarzan. They asked him if he would go ahead 
without the amphibious operation (it will be recalled that the Chinese had 
never been told exactly what sort of operation was contemplated) , or would he 
wait until November 1944 when there might be a major seaborne landing? In 
the meantime, the President suggested, all air transport would be concentrated 
on increasing the tonnage flown to China. Roosevelt and Hopkins held out 
the "fair prospect of terminating the war with Germany by the end of summer 
of 1944," which would release great resources for the Far East, (on the night of 
6 December a poll of the CCS revealed that the earliest date any of them would 
set for the end of the war in Europe was February 1945, with half of them 
guessing it would be spring 1945). 60 

Stilwell's Search for Guidance 

On 6 December Stilwell and his political adviser, John P. Da vies, Jr., met 
with the President and Hopkins. Stilwell had heard of unfavorable develop- 
ments and was anxious to know what effect they would have on U.S. policy in 
China. Thanks to the rapprochement with the Generalissimo in October, the 
American soldier was still joint chief of staff for China Theater, was command- 
ing two divisions of Chinese troops in India and Burma (one of them engaged 
in combat), and was commanding general of the U.S. China, Burma and India 
Theater. The President's radio could be expected to shock the Generalissimo, 
and guidance for Stilwell in the radically changed situation was essential. 

For two years the President's declared policy had been to treat China as a 
Great Power and make of her a partner in a coalition with Britain, Russia, and 
the United States. In the course of this period the President had deferred con- 
tinually to the Generalissimo's wishes, sometimes against the advice of his 

v; (1) William D. Leahy, / Was There: The Personal Story of the Chief of Staff to Presidents Roose- 
velt and Truman Based on His Notes and Diaries Made at the Time (New York: Whittlesey House, 
McGraw Hill Book Company, Inc., 1950) pp. 213-14. (2) Ernest J. King and Walter Muir 
Whitehill, Fleet Admiral King: A Naval Record (New York: W. W. Norton & Company, Inc., 
1952), p. 525. (3) Sherwood, Roosevelt and Hopkins, p. 800. 

60 (1) Sherwood, Roosevelt and Hopkins, pp. 800-801; quotation, p. 801. (2) Arnold, Global 
Mission, p. 473. 



service chiefs. Thus, in March 1943, and again in May 1943, he had overruled 
them to back General Chennault, explaining his decision by the desperate 
urgency of China's need, and the necessity of acknowledging the wishes of the 
Generalissimo as Supreme Commander, China Theater. 

The President had insisted on China's joining in the diplomatic councils of 
the Great Powers and had carried his point just ten weeks before at Moscow. 
In the course of the previous two years the United States had made a number 
of commitments to China, of which the chief was that of Trident, to break 
the blockade of China at the earliest moment. Roosevelt had been a driving 
force in these developments and had often expressed his appreciation of the 
urgent character of China's needs. 

Casablanca, Trident, Quadrant, had erected an imposing structure of 
plans and decisions; an entire new Allied theater, SEAC, under an aggressive 
commander, had been created. All these efforts had seemed to be building to a 
grand climax, Champion, the culmination of these diplomatic and strategic 
efforts. Champion would break the blockade of China, with all the momentous 
ensuing consequences. 

Now, the situation was changed, in a dramatic reversal, and it was essential 
that Stilwell know how the President wanted to meet the situation. The Presi- 
dent explained that the conference had come to an impasse and could not be 
permitted to end in disagreement. Therefore, he would yield to the British point 
of view. The United States and Russia had insisted on Overlord, and so, said 
the President, Churchill had insisted on giving up Tarzan. 61 

So much was clear, and Stilwell asked: "I am interested to know how this 
affects our policy in China." The President's reply was most indefinite. In 
retrospect, it appears that he had not decided what to do about China, and so 
Stilwell could not keep the conversation away from Roosevelt family history, 
the postwar development of China, and the new, postwar Asia. Stilwell and 
Davies prepared minutes of the conversation, and from them, Stilwell tried to 
puzzle out just what the President wanted him to do. 62 

Stilwell concluded that the President's policy was: "Keep China in the war. 
We must retain our flank position [vis-a-vis Japan]. If CKS flops, back 
somebody else." 63 

But how was all this to be done in the face of Buccaneer's cancellation and 
the inevitable compromising of Stil well's position? "Only remarks pertinent to 
question," wrote Stilwell, were "If Tarzan is out, we can boost the [Hump] 
tonnage. VLR bombers [B-29's] can bomb Japan." 64 Several months later 

61 See Story of J. Peene, Sr., atchd illustration. 

62 The Stilwell Papers, page 251, has the text of the conversation. 

63 (1) Story of J. Peene, Sr. (2) Elsewhere, Stilwell gives his impression of the President's 
wishes as: "Poljc^JJtf^e want to help China.'— Period." Stilwell Undated Paper (SUP) 65. (See 
Bibliographical Note.) 

M Story ot J. Peene, Sr. 



Stilwell told Marshall that he had sought guidance at Cairo but had found 
none. 65 Marshall did not challenge this statement. 

Indeed, the President's remarks raised more questions than they answered. 
If, under Japanese attack, or economic distress, the Nationalist regime began to 
crumble, then, according to the President, the United States would "look for 
some other man or group of men, to carry on." 66 Whom did the President 
have in mind, a dissident war lord like Marshal Li Chi-shen or the Commu- 
nists? At what point was Stilwell to begin dealing with such people? 

Knowing that Stilwell's position in China would be almost impossible after 
Sextant, Marshall offered him a high post in another theater. Stilwell declined 
it. 67 Talking with Marshall the day after his interview with the President, 
Stilwell learned: "George hopeful about Germany. 'Hang on and keep going.' 
Nothing else he could tell me. Everything dangling." 68 

One thing, arms for ninety divisions, might have kept the Generalissimo 
from regarding Sextant as an utter disappointment. On 10 December Stilwell 
attended a meeting to discuss the project. 69 Three weeks later, after Stilwell 
returned to CBI, Marshall was told by OPD: "The commitment regarding the 
Lend-Lease equipping of Chinese divisions the President actually made at 
Sextant is not known. We are proceeding on the assumption the President 
made no commitment on the timing of the flow of equipment." 70 Stilwell was 
informed accordingly. As for the landing craft that on Churchill's insistence 
were taken from SEAC to reinforce Anvil, several weeks later the British Chiefs 
of Staff, supported by the Prime Minister, made the first of several attempts to 
have Anvil canceled for operations elsewhere in the Mediterranean. 71 

The Generalissimo's answer to the President's radio telling him of 
Buccaneer's cancellation was awaited anxiously, for SEAC could have no CCS 
directive on amphibious operations until it was known how the Chinese would 
react to the disappointment. Discussion of future operations continued while 
the CCS awaited his reply. General Marshall suggested that the land operations 
outlined by SEAC might well be undertaken by the Chinese advancing from 
Yunnan and screened by the U.S. long-range penetration groups directed at 
Quebec, with some of the troops released by Buccaneer forming a reserve. 
The Chief of the Imperial General Staff countered with the proposal that 
Mountbatten's new mission should be to guard Assam by active offensive 
operations. 72 

65 CM-IN 4651, Stilwell to Marshall, 7 Mar 44. 

66 The Stilwell Papers, p. 252, quoting Roosevelt. 
67 Interv with Marshall, 6 Jul 49. 

68 Stilwell Diary, 7 Dec 43. 

69 Stilwell Diary, 10 Dec 43. 

70 Memo, Handy for Marshall, 31 Dec 43, sub: Equipping Chinese Divs, sent as CM-OUT 
11706, Marshall to Sultan for Stilwell, 31 Dec 43. 

71 Ltr, Lt Gen Frederick E. Morgan, COSSAC, to Secy, COS, 6 Jan 44. COSSAC (44) 5, AFHQ 
G-3 File, OCMH. Also published as CCS 446/1, 8 Jan 44, sub: Three Div Lift for Anvil. ABC 
384 (Europe) 1 Mar 43, Sec 2A, A48-224. 

72 Min, CCS 138th Mtg, 7 Dec 43. 



Meanwhile, Stilwell sent a radio to General Hearn in Chungking, ordering 
Hearn to see the Generalissimo and urge him to go ahead with his share of the 
campaign, regardless of Buccaneer's cancellation. 73 This action was consistent 
with Stilwell's often expressed view that seizure of the Andaman Islands con- 
tributed nothing to operations in Burma. His reasoning was supported by the 
facts that the Japanese had opened a railway to Thailand, so that they no longer 
depended on the port of Rangoon, and that airfields on the Andamans were 
only 100 miles closer to Rangoon than those already in Allied hands, so that 
their possession would not be decisive in air operations against Rangoon, even 
if such were of vital importance. 

When the Generalissimo's answer to the President arrived at Cairo on 
9 December, it spoke in ominous tones: 

I have received your telegram of December Sixth. Upon my return I asked Madame 
Chiang to inform you of the gratifying effect the communique of the Cairo Conference has 
had on the Chinese army and people in uplifting their morale to continue active resistance 
against Japan. This letter is on the way and is being brought to you by the pilot, Captain 

First, prior to the Cairo Conference there had been disturbing elements voicing their 
discontent and uncertainty of America and Great Britain's attitude in waging a global war 
and at the same time leaving China to shift as best she could against our common enemy. 
At one stroke the Cairo communique decisively swept away this suspicion in that we three 
had jointly and publicly pledged to launch a joint all-out offensive in the Pacific. 

Second, if it should now be known to the Chinese army and people that a radical change 
of policy and strategy is being contemplated, the repercussions would be so disheartening 
that I fear of the consequences of China's inability to hold out much longer. 

Third, I am aware and appreciate your being influenced by the probable tremendous 
advantages to be reaped by China as well as by the United Nations as a whole in speedily 
defeating Germany first. For the victory of one theater of war necessarily affects all other 
theaters; on the other hand, the collapse of the China theater would have equally grave 
consequences on the global war. I have therefore come to this conclusion that in order to 
save this grave situation, I am inclined to accept your recommendation. You will doubtless 
realize that in so doing my task in rallying the nation to continue resistance is being made 
infinitely more difficult. 

(1) Because the danger to the China theater lies not only in the inferiority of our military 
strength, but also, and more especially, in our critical economic condition which may seri- 
ously affect the morale of the army and people, and cause at any moment a sudden collapse 
of the entire front. Judging from the present critical situation, military as well as economic, 
it would be impossible for us to hold on for six months, and a fortiori to wait till November 
1944. In my last conversation with you I stated that China's economic situation was more 
critical than the military. The only seeming solution is to assure the Chinese people and 
army of your sincere concern in the China theater of war by assisting China to hold on with 
a billion gold dollar loan to strengthen her economic front and relieve her dire economic 
needs. Simultaneously, in order to prove our resolute determination to bring relentless pres- 
sure on Japan, the Chinese air force and the American air force stationed in China should be 
increased, as from next spring, by at least double the number of aircraft already agreed upon, 
and the total of air transportation should be increased, as from February of next year, to at 
least 20,000 tons a month to make effective the operation of the additional planes. 

73 Rad AMSME 1720, Stilwell to Hearn, 7 Dec 43. Item 1502, Bk 5, JWS Personal File. 



(2) In this way it might be possible to bring relief to our economic condition for the 
coming year, and to maintain the morale of the army and the people who would be greatly 
encouraged by America's timely assistance. What I have suggested is, I believe, the only way 
of remedying the drawbacks of the strategy concerning the China and Pacific theaters. I am 
sure you will appreciate my difficult position and give me the necessary assistance. I have 
instructed General Stilwell to return immediately to Chungking and I shall discuss with him 
regarding the details of the proposed changed plan and shall let you know of my decision 
as to which one of your suggestions is the more feasible. 

From the declaration of the Teheran Conference Japan will rightly deduce that practically 
the entire weight of the United Nations' forces will be applied to the European front thus 
abandoning the China theater to the mercy of Japan's mechanized air and land forces. It 
would be strategic on Japan's part to (3) liquidate the China Affair during the coming year. 
It may therefore be expected that the Japanese will before long launch an all-out offensive 
against China so as to remove the threat to their rear, and thus re-capture the militarists' 
waning popularity and bolster their fighting morale in the Pacific. This is the problem which 
I have to face. Knowing that you are a realist, and as your loyal colleague, I feel constrained 
to acquaint you with the above facts. Awaiting an early reply, 

Chiang Kai-shek 74 

The Generalissimo's requests were not enough to bring agreement on a new 
directive to SEAC for a major amphibious operation. For the time being SEAC 
and Stilwell would have to be governed by the Sextant decisions, which were 
sufficiently explicit. These ordered the occupation of upper Burma in spring 
1944 (1) to improve the air route and (2) to open land communications with 
China. An amphibious operation at the same time was approved. Twilight, 
the B-29 project, was also approved, and the Fourteenth Air Force, the Chinese 
Army, and the Chinese Air Force would be improved for intensified operations 
in and from China. The general concept of the Sextant decisions on the 
Pacific and Asia was that "the main effort against Japan should be made in the 
Pacific." What was attempted elsewhere in Asia would be in support of that 
main effort. There would be first priority for Anvil and Overlord, the 
supreme operations for 1944. 75 

SEAC Tries To Salvage Burma Operations 

Admiral Mountbatten was an aggressive commander, of proven desire to 
close with the enemy. Moreover, he and his subordinates, of whom Stilwell was 
one, were bound by the Sextant decision to clear north Burma. Lastly, fight- 
ing in the Arakan and in north Burma had been under way for weeks, with 
both sides reinforcing. Buccaneer's demise left SEAC the alternatives of post- 
poning an attempt at a major co-ordinated offensive for another year, which 
would probably mean the end of operations to clear north Burma, or of staging 
an amphibious operation smaller than Buccaneer, with the hope that it would 

74 Rad AGWAR 919, Chiang to Roosevelt, 9 Dec 43. Item 1505A, Bk 5, JWS Personal File. 

" (1) CCS 417, 2 Dec 43, sub: Plan for Defeat of Japan. (2) CCS 426/1, 6 Dec 43, sub: Rpt 
to President and Prime Minister. (3) CCS 397 (rev), 3 Dec 43, sub: Specific Opns for Defeat of 



still be enough to meet the Generalissimo's stipulation for such an operation 
in the Bay of Bengal, and so lead him to take active part in the Burma fighting. 

Mountbatten's first reaction was hesitant, because the shipping requirement 
would be the same if the attempt was large or small, and because no worth- 
while objective could be seized with what shipping was at hand. When an 
amphibious assault on the Arakan coast was first proposed, he did not see how 
it could be presented as that previously promised the Generalissimo or how it 
alone could fulfill SEAC's basic directive. However, since such would be a 
starting point for the future, would enable the long-range penetration groups 
to do their work, and would not commit him to an offensive in central Burma, 
he directed his staff to study it. 76 

Since the amphibious operation promised the Generalissimo had never been 
defined to him, and since his stipulation had been for a major one, if the SEAC 
planners could somehow evolve a major amphibious effort the question of 
Allied good faith would be answered, even if belatedly, and attention would be 
focused on the Generalissimo's reaction. By adjusting the delicate balances for 
a plan that might be imposing enough to satisfy the Generalissimo yet still fit 
within SEAC resources, SEAC's planners evolved Pigstick. Pigstick called 
for an assault on the Mayu peninsula aimed at Akyab. Two divisions plus two 
brigades would be used in a southward advance down the peninsula and one 
division in an amphibious assault aimed at surrounding and destroying not less 
than 20,000 Japanese. One more landing like Pigstick, perhaps in the Ramree- 
Cheduba area, could take staging areas that would put 15 Corps within reach 
of Rangoon. 77 Tarzan was modified into Gripfast, an attack on north and 
central Burma with an airborne landing at Indaw on the Japanese line of 
communications to Myitkyina. 

In the initial negotiations between Mountbatten and the Chinese on the 
commitment to battle of the U.S.-sponsored Chinese divisions in Yunnan 
(Y-Force) Stilwell entered enthusiastically. SEAC's new plan, thought Stilwell, 
was almost the same as Toreador (the airborne landing in central Burma), 
which had so appealed to the Generalissimo at Sextant. 78 Mandalay itself was 
now the objective of SEACs efforts, while the amphibious operations were 
enlarged. 79 

For whatever reasons, the Generalissimo was unimpressed with SEAC's 
attempt to meet his demands for an amphibious operation before he would 
move. Like a wary customer, he questioned the value of the substitute that 
SEAC was offering. Since even in the genial atmosphere of Cairo he had been 
conspicuously unwilling to commit himself, it was apparent that he would 
drive a hard bargain, particularly since the President's radio from Cairo had 
offered him an alibi. His final reply to the President's radio on 17 December 

76 (1) Se J Ch. I J above. (2) Rad, CCS to Mountbatten, 5 Dec 43; Rad, Wedemeyer to Mount- 
batten, 6 Dec 43; Rad, Mountbatten to COS, 11 Dec 43. SEAC War Diary. 

77 Rad, SEAC (RL) 19, 19 Dec 43. ABC 384 (Burma) 8-25-42, Sec IV, A48-224. 

78 Stilwell Black Book, 19 Dec 43. 

79 Memo, Stilwell for Generalissimo, 19 Dec 43. Item 1533, Bk 5, JWS Personal File. 



stressed his need for money and air power but implied that a large enough 
amphibious operation might even yet secure his co-operation. 

My telegram of December 10th must have reached you by this time. I have discussed 
with General Stilwell the proposed change in the plan of campaign and have come to the 
following conclusions: 

In case the original plan of concentrating warships and transports for landing troops 
cannot be completely carried out, it would be better to defer the amphibious all-out offensive 
till November next as you suggested so that the enemy in Burma may be annihilated once 
and for all. In the meantime preparations for an offensive against Burma next spring should 
proceed at full speed as originally planned, thus enabling us to launch an attack on land at 
any moment which is deemed favorable, or at any time before next autumn if a sufficient 
number of warships and transports can be concentrated to effect a grand scale landing on the 
enemy's flanks, without waiting till the autumn of next year. 

In this way the Burma front might be liquidated sooner than one could anticipate. I have 
decided to accept your suggestion that the general offensive against Burma should be post- 
poned to November next or sooner if the original amphibious operation could be launched. 
At the same time I cannot but reiterate that in the intervening period of one year during 
which there will be little hope of re-opening the Burma Road, the China theater of war will 
be in a most critical situation. I therefore earnestly ask you to do all in your power to accede 
to my request for financial assistance and for an increase of air force and air transportation as 
stated in my telegram of December 10th, in the hope that the danger to the China theater 
may be removed and the drawbacks in the strategy against Japan remedied in accordance 
with your consistent friendly policy of rendering assistance to China. Awaiting an early reply. 

Chiang Kai-shek 80 

Doing his best to meet the Generalissimo's requirements, Mountbatten 
gave Stilwell for a further "talking point" information that the Chindit forces 
in the proposed Burma operations would total 20,000 men, approximately half 
of whom would be assisting the Chinese advance. 81 

In talks with the Generalissimo and Madame Chiang, Stilwell learned that 
the Chinese expected the United States to pay the entire cost of constructing 
the B-29 fields at Cheng-tu. The Generalissimo's request for a loan of one 
billion dollars gold, the Chinese insistence on setting an official exchange rate 
of 20 to 1 between their currency and the U.S. dollar when the black market 
rate was 240 to 1, and rising rapidly, and now the President's alleged promise 
to pay the whole cost of the B-29 fields introduced a new factor of importance, 
the sheer monetary cost of attempting operations in China. 

The Generalissimo estimated that the Cheng-tu fields would cost two to 
three billion dollars of Chinese currency. "At 20 to 1, at least 100 million gold, 
of which one-half will be squeeze. Appalling," wrote Stilwell. Stilwell protested 
that his understanding was the United States would "help" with the project. 
No, retorted Madame Chiang, the President had promised to pay for every- 
thing. Disgusted by what to him seemed a naive softness, Stilwell wrote: "One 
more example of the stupid spirit of concession that proves to them that we are 
suckers. 'We'll put in VLR bombers' (no bargaining). Then, 'we'll pay for the 

80 Rad AG WAR 941, Chiang to Roosevelt. 17 Dec 43. Item 1529, Bk 5, J WS "Personal File. 

81 Rad COPIR 10, Mountbatten to Stilwell, 20 Dec 43. Item 1541, Bk 5, JWS Personal File. 



fields' (no bargaining). Same on air freight— promise without bargain. Same on 
equipment for army— promise without bargain. Same on Chinese Air Force. 
Same on 14th Air Force. Same on everything." 82 

When the discussion came around to the current operations in Burma, the 
Generalissimo's actions on 18 and 19 December baffled Stilwell. On the 18th 
the Generalissimo gave Stilwell full command of the Chinese forces in India 
and those now fighting in the Hukawng Valley. The next day he rejected 
Mountbatten's proposals for a major attack on Burma, which made Stilwell 
write: "[The Generalissimo] is afraid that even concerted attack by all available 
forces has only one chance in a hundred and yet he'll sit back and let a small 
force take on the Japs alone." 83 With the Generalissimo's promise in hand, 
Stilwell prepared to leave to take command of the Hukawng Valley operations. 
He believed that with the Ledo Force there was just a chance he might be able 
to link with the Chinese Yunnan divisions somewhere near Myitkyina. 84 

Stilwell's decision to assume active command of the forces in north Burma 
is not discussed or analyzed in his private or official papers. In the light of his 
habit of analyzing every major step this circumstance suggests he thought the 
move an obvious one. By December 1943 the post of chief of staff to the 
Supreme Commander, China Theater, was simply a paper one, without staff, 
directives, or duties. The Chinese had never agreed to set up the Sino- American 
staff through which Stilwell was to have functioned as Joint Chief of Staff, 
China Theater. After the Three Demands crisis of June 1942 the Generalissimo 
had largely ignored both him and his suggestions. Therefore, Stilwell's post of 
chief of staff to the Generalissimo would not require his presence in China. 

There was Stilwell's still-existing mission of improving the combat efficiency 
of the Chinese Army, but his superiors had not objected to his conclusion that 
because Chinese delay had wasted two years there was little more he could do, 
and were themselves coming to the very similar conclusion that little more 
should be attempted than that which Stilwell had already begun, and which his 
subordinates in China could carry out as a matter of routine. 

Since October 1943 the only major development had been the Sextant 
Conference, which had so obviously compromised Stilwell's position in China 
that Marshall had asked him if he wanted to be recalled. Mountbatten, Stilwell's 
superior, was actively soliciting the Generalissimo's aid in Burma operations, 
thus relieving Stilwell of responsibility for that task. 

There remained the operations in Burma, which had been under way since 
30 October 1943. For two months the American officers of Chih Hui Pu had 
been trying to achieve a satisfactory solution, but without success. General 
Boatner, Stilwell's deputy in north Burma, who had been actively exercising 
field command, was now a victim of pneumonia. 85 General Sun, who might 

82 Stilwell Black Book, 18 Dec 43. 

83 (1) Stilwell Diary, 18 Dec 43. (2) Stilwell Black Book, 19 Dec 43. (3) Quotation from The . 
Stilwell Papers, p. 265. 

84 The Stilwell Papers, p. 266. 

85 Stilwell Diary, 21 Dec 43. 



have commanded, had made it very plain that he wanted to retreat. The Chinese 
were now heavily engaged, and the situation had been described to Stilwell as 
critical. So Stilwell prepared to go to north Burma and assume command in the 
jungle. He was then sixty years of age. 86 

The conduct of American military-diplomatic relations with China was 
tacitly assumed by the President. In 1942 and 1943 Stilwell had presented many 
memorandums to the Generalissimo, to which the Chinese had rarely replied. 
In 1944, the President sent one message after another to the Generalissimo on 
military matters, and these the Generalissimo could not ignore. As will be seen, 
the role of CBI Theater headquarters in these exchanges was the humble and 
mechanical one of delivering the text of these presidential proddings to the 

A Changing U.S. Attitude 

Once again in the history of the U.S. effort in China, Burma, and India, the 
issues were about to be placed before the President, this time by Stilwell at 
Madame Chiang's suggestion. Stilwell was not hopeful of the President's will- 
ingness to intercede, but he adopted the suggestion. 87 Manifestly, Stilwell did 
not feel that the action of the President and the Prime Minister in reneging at 
Cairo on the long-promised amphibious operation made it unnecessary or 
inadvisable for the Generalissimo to take action in Burma or that it made 
ungraceful any criticism of the Generalissimo's reluctance from within those 
powers that had broken their pledges to him. So, Stilwell told Marshall that the 
SEAC plan was now virtually what CBI Theater had been urging all along, that 
if the President would exhort the Generalissimo to cross the Salween River 
when his allies attacked Burma, the Chinese leader might play his part. 88 If the 
Generalissimo knew of this move, he could have reflected that his own message 
to the President two days before had accepted one of the two choices the Presi- 
dent had offered, and that in the past the President had extended credits, 
lend-lease, and air support without asking anything in return. 

Drafted by the War Department, the President's reply indicated that 
Roosevelt had moved away from the Generalissimo's and Chennault's views 
and was a great deal closer to Stilwell's. The President returned a qualified 
negative to the Generalissimo's requests. Describing himself as fully aware of 
the military and economic situation in China, the President said that the best 
the United States could do was to aid in the immediate opening of a land line 
of communications to China. The military actions involved in so doing would 
afford greater protection to the Hump air route. Roosevelt told the General- 
issimo of Mountbatten's planning the largest possible operation to retake 
Burma and expressed his hope that the Generalissimo would do everything he 

86 The Stilwell Papers, p. 285. Stilwell's sixty-first birthday was on 20 March 1944. 

87 (1) Stilwell Black Book, 19 Dec 43. (2) The Stilwell Papers, p. 263. 

88 Rad AGWAR 947, Stilwell to Marshall, 19 Dec 43. Item 1537, Bk 5, JWS Personal File. 



could to carry out the part reserved for China. Nothing whatever was said about 
postponing active operations until November 1944. Roosevelt discounted what 
could be done by more air power in China until the line of communications 
had been improved. The Chinese might find comfort in the President's assur- 
ance that plans to increase Hump capacity to 12,000 tons a month were well 
advanced, provided an advance by the Allied ground forces forestalled a 
Japanese attempt to interrupt the airline. The message closed with the brief 
comment that the Treasury Department was weighing China's request for a 
billion-dollar (U.S.) loan. 89 

Thus, the President was suggesting that China act and was stressing action 
on the ground rather than in the air. The Generalissimo had accepted one of 
the alternate courses offered by the President, waiting until November 1944 to 
advance into Burma, only to find that the President had quietly abandoned it. 
Did the Generalissimo's linking the cancellation of Buccaneer with a request 
for one billion dollars anger the President? Whatever the reason, the changed 
tone and shifts in emphasis of the President's reply, the ever stronger and more 
demanding nature of its successors, suggest that the President had made up his 
mind about China. At Cairo Roosevelt had been uncertain and unable to guide 
Stilwell; after Cairo and a few weeks of consideration, the President was striking 
out along the line of insisting that China take the offensive in return for the 
lend-lease she had received. 

Despite the President's urgings, the Generalissimo's reply was negative. It 
even had overtones of the sardonic. He agreed to leave the Ledo forces at 
Mountbatten's disposal but stated that the Y-Force would move only if the 
Allies took the Andaman Islands, Rangoon, or Moulmein. If they succeeded in 
taking Mandalay or Lashio, he would order his armies into Burma even if there 
was no amphibious operation. 90 

General Hearn, to whom Stilwell had entrusted the American share of 
negotiations with the Generalissimo, did not believe the Generalissimo's reply 
was final but thought rather that he was bargaining for a bigger amphibious 
operation or a pledge that the Burma campaign would definitely include cap- 
ture of Rangoon. Nor did he believe the Generalissimo was aware of the size 
of the effort that Mountbatten might be able to make. If the Generalissimo 
agreed to commit Y-Force, 325,000 Allied combat troops would be involved in 
the Burma operation. 91 

Though urged by Hearn and Stilwell to accept the Generalissimo's 
Mandalay-Lashio offer, this was further than Mountbatten would go. Indeed, a 
certain asperity was entering his references to the Chinese. Asking that the 

89 (1) Item 58, OPD Exec 10. (2) Rad WAR 4092, Roosevelt to Chiang, 20 Dec 43. Item 
1546, Bk 5, JWS Personal File. 

90 ( 1) Rad, Lt Gen Sir Adrian Carton de Wiart, Prime Minister's and SAC's Personal Repre- 
sentative to Chungking, to Mountbatten, 23 Dec 43. SEAC War Diary. (2) Rad AM 2934, Hearn 
to Merrill, 28 Dec 43; Rad AM 2372, Sultan to Stilwell, 30 Dec 43. Items 1571, 1587, Bk 5 JWS 
Personal File. (3) CM-IN 1161, Hearn to Marshall, Handy, and Maj Gen Joseph T. McNarney, 
2 Jan 44. 

91 CM-IN 14577, Hearn to Stilwell and Marshall, 23 Dec 43. 



United States put pressure on the Generalissimo, Mountbatten remarked: "I do 
not see why we should continue to supply him with munitions if they are to be 
used solely for internal political purposes." 92 

Still determined on an offensive, Admiral Mountbatten went on with 
preparations for Pigstick, the assault on the Mayu peninsula. He told the 
British Chiefs of Staff that while Pigstick was within SEAC's capabilities, "if 
any further resources are taken from me ... I shall have to cancel the 
operation." 93 

That the British Chiefs of Staff did not favor Pigstick became apparent 
when they suggested to the CCS that if Pigstick was canceled three fast LST's 
(landing ship, tank) and other landing craft could be released for a landing at 
Anzio, Italy. After examining the landing craft situation in the Mediterranean 
and considering the old promise to the Generalissimo to make an amphibious 
operation, the Joint Chiefs of Staff urged that plans and preparations for Pig- 
stick continue with no further withdrawal of landing craft from SEAC. More- 
over, Stilwell's initial attempts to advance in north Burma were meeting with 
success and an operation to the south would divert some Japanese from him. 

While these discussions between the Joint and British Chiefs of Staff were 
under way, the British Chiefs of Staff told Mountbatten that they did not think 
Pigstick could be carried out, and, although there was still no decision by the 
CCS, ordered him to return the landing craft in question to the Mediterranean. 
The departure of the craft, together with the warning by his commanders in 
chief that they could not carry out Pigstick during the favorable weather 
period of February 1944 unless it was ordered by 30 December at the latest, 
forced Mountbatten to cancel the operation without awaiting CCS approval. 94 

Admiral Mountbatten canceled Pigstick with reluctance, for the action 
meant to him the probability of no worthwhile offensive against the Japanese 
for at least a year after SEAC's formation and would have an adverse effect on 
morale. In a last attempt at an amphibious operation, Mountbatten ordered 
preparations for Bulldozer, a much smaller amphibious operation in the 
Arakan. A message from Mr. Churchill to "mark time for a day or two till we 
get matters cleaned up" was enough to end it, for even a day's delay would 
affect the time to mount it before rough weather began in the Bay of Bengal. 
Thus, the last hope of meeting the Generalissimo's demand for an amphibious 
operation was gone. 

These events were enough to dampen even the buoyant Mountbatten's 
enthusiasm for a Burma campaign. Where a week before he had said: "I have 
no intention of allowing operations in Northern Burma to fade on account of 
abandonment of proposed operations elsewhere," he now told his staff: "The 

92 Rad SEACOS 53, Mountbatten to COS, 24 Dec 43. SEAC War Diary. 

93 Rad SEACOS 54, Mountbatten to COS, 27 Dec 43; Min, SAC's 37th Mtg, 27 Dec 43. SEAC 
War Diary. 

94 (1) CCS 452, 30 Dec 43, sub: Cancellation of Opn Pigstick. (2) CCS 452/2, 6 Jan 44, sub: 
Cancellation of Opn Pigstick. (3) Rad, COS to Mountbatten, 29 Dec 43; Rad, Mountbatten to 
CCS, 6 Jan 44. SEAC War Diary. 



quickest and most efficient way of taking supplies on a large scale into China 
is through a port rather than by a long and uncertain land route." 95 

Reflecting the strategic developments of Sextant and the Generalissimo's 
reluctance to engage in Burma operations, the Strategy and Policy Group, 
OPD, on 8 January 1944 submitted its comments on the "future military value 
of China Theater." The planners stated that since the main effort in the Pacific 
would be made in the central and southwest areas of that great expanse, the 
mission of Stilwell's CBI Theater should be to give air support to the main 
effort. The bases from which this support was to come should be in areas 
already secure, because to acquire any more territory would require of the 
Chinese Army an efficiency not likely to be attained before 1946-47. No further 
effort should be made, the paper went on, to equip Chinese ground forces 
beyond enabling them to control areas they already had. Therefore, all avail- 
able Hump airlift capacity should be devoted to building up air power in 
China, which was believed to be the best way of preventing China's collapse, 
as well as of aiding Pacific operations. Offensive operations in Burma to thwart 
a Japanese threat to the existing India-China air line of communications were 
still thought necessary. 96 


Before the Sextant Conference, the United States placed great emphasis on 
major operations in Southeast Asia to break the blockade of China and divert 
Japanese strength from the Southwest Pacific. President Roosevelt had been 
most interested in the implications of this policy as it applied to Asia. At Sex- 
tant his attitude changed; the amphibious operation demanded by the Gener- 
alissimo as the price of his co-operation in Burma was canceled, and for a time 
it seemed the President was willing to postpone Burma operations until 
November 1944. The Generalissimo asked for a billion U.S. dollars and heavy 
air reinforcements so that China might withstand another year's blockade. He 
was not willing himself to make a major effort to break it. The President's 
reply was drafted by the War Department and moved toward full support of 
Stilwell. During these discussions, the British Chiefs of Staff withdrew certain 
essential landing craft from Mountbatten, in effect ending his hopes of a major 
amphibious operation. December ended with Stilwell taking his post in north 
Burma to command the now heavily engaged Chinese New First Army, with 
the President urging China to play a more active part in the war, and with 
OPD suggesting that the mission of CBI Theater should be to give air support 
to Allied operations in the Pacific. 

^ (1) Rad, Churchill to Mountbatten, 7 Jan 44; Extract, SAC's Personal Diary, 28 Dec 43; 
Quotation from Min, SAC's Fifth Stf Mtg, 6 Jan 44. SEAC War Diary. (2) JPS 346, 2 Jan 44, sub: 
Cancellation of Opn Pigstick. (3) Notes by Brig. Gen. Frank N. Roberts on draft manuscript of 
this chapter. OCMH. 

96 Memo, Gen Roberts, Chief, Strategy and Policy Gp, OPD, for ACofS OPD, 8 Jan 44, sub: 
Future Mil Value of China Theater; Memo, Billo for Roberts, 13 Jan 44, sub: Future Mil Value of 
China Theater. OPD 201 (Wedemeyer, A. C), A47-30. 


The India-Based Air Effort 

On 21 December 1943, Stilwell arrived at Ledo to take over the North 
Burma Campaign, "under better auspices than last time." 1 No student of 
classic antiquity, Stilwell probably did not have in mind the full definition: 
"aus'pice. ... 1. A sign, . . . used in augury, as the flight of birds, ... or 
phenomena in the sky." But if the omens were better in December 1943 than 
they had been in March 1942, it was largely because of happenings in the skies 
over India and Burma. Anglo-American air power, based on India, was ap- 
proaching giant maturity, and was about to make possible a campaign of a type 
never seen before, a campaign in which the customary Japanese tactic of en- 
circlement was turned against its authors with devastating results. To under- 
stand the differences between the campaigns of 1942 and 1943-44, one must 
understand techniques of air support and air supply evolved between March 
1942 and December 1943, and the extent of Allied air superiority in Burma. 

Command and Administration of the Air Forces 

After Sextant, Mountbatten reached a firm decision that his command 
must include an integrated (i.e., Anglo-American) air force. His earlier hints 
of such a move to Marshall, Arnold, and the Generalissimo had been well re- 
ceived. He had been impressed by the work of an integrated air force in North 
Africa, and felt he could not "accept a state of affairs where a subordinate Com- 
mander in my theatre had independent responsibilities for combat air oper- 
ations." On learning of the project, Stilwell and General Stratemeyer, 
commanding the Army Air Forces, India-Burma Sector, objected, for they be- 
lieved that if there was to be no operation for retaking all Burma, and at this 
time none seemed very likely, there was no need of an integrated air force. In 
a way, this was a reversal by Stratemeyer, who had first rather liked the idea, 
then had changed his mind after Mountbatten's October visit to Chungking. 

Despite the opposition of the two senior U.S. commanders in SEAC, 
Mountbatten thought it time to establish his authority in the theater and felt 
sure that the CCS would not reverse him. Marshall approved, though he re- 

1 The Stilwell Papers, p. 267. 



minded Mountbatten of the U.S. commitments to China and warned that the 
time would come when the U.S. would have to move part of the Tenth Air 
Force to China. 2 

On 12 December the Tenth Air Force and the Royal Air Force (RAF) 
Bengal Command joined their combat strength to become the Eastern Air 
Command (EAC), though preserving their administrative entities. Stratemeyer 
assumed command 1 5 December. He had the Strategic Air Force under opera- 
tional control of Brig. Gen. Howard C. Davidson, the Tenth Air Force's com- 
mander; the Third Tactical Air Force under operational control of Air Marshal 
Sir John Baldwin of Bengal Command; the Troop Carrier Command under 
operational control of Brig. Gen. William D. Old, and the Photographic Re- 
connaissance Group under Group Captain S. C. Wise. Third Tactical Air Force 
was in turn composed of Northern Sector Air Force (AAF) in north Burma, 
221 Group (RAF) in Manipur, and 224 Group (RAF) in the Arakan. Naming 
Stratemeyer to command was a graceful gesture by Mountbatten, for the bulk 
of EAC's aircraft, especially fighters, were British. Eastern Air Command began 
operations in January 1944 with 576 fighters (141 AAF), 79 heavy bombers 
(48 AAF), 70 medium bombers (37 AAF), 10 AAF reconnaissance aircraft, and 
84 transports (51 AAF). 3 This strength was cut a fortnight later when the 34lst 
Bombardment Group (M), less the 490th Squadron, was assigned to the Four- 
teenth Air Force in China. 

In June 1944 the combat strength of the Tenth Air Force was reconstituted 
as a tactical air force of two fighter groups and one troop carrier group plus 
administrative and service elements to support the North Burma Campaign. 
Experience had demonstrated that the North Burma Campaign was tactically 
independent of operations in Manipur State and the Arakan and that oper- 
ations in the three sectors did not require central direction of tactical air sup- 
port. Troop Carrier Command was eliminated and its units split between Tenth 
Air Force and Third Tactical Air Force. This then gave EAC five subordinate 
air commands: Strategic Air Force (still an integrated Anglo-American force); 
Third Tactical Air Force (mostly RAF plus the 12th Bombardment Group [M] 
and the 459th Fighter Squadron); the Photographic Reconnaissance Group 
(Anglo-American); the Tenth Air Force; and 293 Wing RAF, charged with de- 
fense of the Calcutta industrial area. 4 

2 (1) Quotation from Extract, S AC's Personal Diary, 11 Dec 43. SEAC War Diary. (2) 
Stilwell's Mission to China l Ch. V111.K 3) Ltr, Mountbatten to Arnold, 27 Nov 43; Ltr, Marshall to 
Mountbatten, 4 Jan 44. Item 676, Msg Bk 18, OPD Exec 9. 

3 Lt. Gen. George E. Stratemeyer, Despatch on Air Operations in Eastern Air Command (SEA) 
Covering the Period 15 December 1943 to 1 June 1945, MS (hereafter EAC Despatch), App. 16, 
Incls 1, 2. OCMH. The heavy bombers were B-24's and Wellingtons with a few Lancasters and 
Halifaxes; the mediums were B-25's; the fighters were mostly Hurricane He's, lid's, and P-40's. 
Five squadrons were so fortunate as to have Spitfires, while two others had P-38's and P-51's. The 
Vultee Vengeance dive bomber was used by five squadrons, Burma being one of the few areas in 
which it was to be found. The versatile and successful Beaufighters acted as night fighters and light 

4 EAC Despatch, pp. 44, 7-8. 



The Japanese Air Effort in Burma 

The senior headquarters for the Japanese Army air force in Southeast Asia 
was 3d Air Army at Singapore. The defense of Burma was still charged to the 
5th Air Division, veterans of the First Burma Campaign. Under 5th Air Division, 
in late November 1943, were the 4th Air Brigade, with the 8th (light bomber), 
34th (light bomber), 77th (fighter), and the 50th (fighter) Air Regiments; and 
the 7th Air Brigade, with the 12th ("heavy" bomber), 98th ("heavy" bomber), 
and 64th (fighter) Air Regiments. Also under 5th Air Division were the 21st, 33d, 
17th, and 204th Air Regiments (fighter), and the 81st Air Regiment (reconnais- 
sance). After the war, the Japanese estimated this force comprised about 129 
fighters, 30 "heavy" bombers, 47 light bombers, and 15 reconnaissance aircraft. 
The Japanese naval air force was also represented. It had been drained by the 
fighting in the Southwest Pacific, but at least its 28th Air Regiment was present 
in October 1943, with 30 fighters and 9 bombers. 5 

Eastern Air Command estimated Japanese strength at 277 aircraft, including 
reconnaissance craft. Though outnumbered by EAC, 5th Air Division had a few 
counterbalancing advantages. Its rear areas were well out of the reach of all but 
the heavy bombers, which were too few to neutralize Japanese bases. The 5th 
Air Division had scores of forward airstrips so that it could prepare its oper- 
ations in the rear areas, stage them through any of the forward strips, then 
quickly disperse on returning. Pilots and aircrews were good; their equipment, 
adequate. 6 

Not waiting for their opponents, the Japanese airmen undertook to cripple 
the anticipated Allied attack on Burma by genuine strategic bombing, a series 
of attacks on Calcutta. The fanfare attendant on SEAC's creation, the appoint- 
ment of an admiral as Supreme Allied Commander, and signs of Allied prep- 
arations around Ledo, Imphal, and Calcutta convinced the Japanese that a 
major operation against Burma was imminent. 7 

Neutralization of Calcutta seemed advisable. Preparations began in October 
1943. The 4th Air Brigade attacked RAF fields about Imphal in November to 
divert attention and resources from the Calcutta area. Calcutta itself was at- 
tacked on 5 December by the 7th Air Brigade, with 94 fighters and 18 heavy 
bombers from six air regiments, and 8 reconnaissance aircraft, plus 30 fighters 
and 9 bombers of the Imperial Navy's 28th Air Regiment. "Considerable damage 

5 (1) The Japanese fighters were mostly single-engined 1941 model Oscars, with two 12.7-mm. 
guns and with a top speed of 334 mph at 16,400 feet, and twin-engined two-seated 1942-model 
Nicks, with three 12.7-mm. and one 20-mm. guns and with a top speed of 340 mph at 13,000 feet. 
The light bomber was the Lily, with a 57-foot wingspread and a 1,100-pound bomb load. The 
"heavy" bombers were Sallys, with a maximum bomb load of 2,200 lbs. and with a wingspread of 
70 feet. The Sally was a 1937 model; the Lily was a 1939 model. TM-E 30-480, Handbook on 
Japanese Military Forces, 15 September 1944, pp. 58-75. Japanese Officers' Comments, p. 8. (2) 
Japanese Study 94. (3) SEATIC Bull 248, 22 Apr 47, pp. 24-25. MID Library. 

6 EAC Despatch, pp. 12-13. 

7 (1) Japanese Studies 89 and 94. (2) SEATIC Bull 242, 7 Aug 46, p. 12; SEATIC Bull 247, 
22 Apr 47, pp. 110-11. MID Library. 



was done to dock area," Eastern Air Command admitted later, and its dispatch 
adds: "The Japanese Air Force, in fact, controlled the air over Burma and, while 
maintaining a constant threat against the vital air route to China, was harassing 
Allied ground and air installations in the Assam valley, the Arakan, and 
Bengal." 8 

Though a second attempt at the strategic bombardment of India was thus 
launched by the Japanese, and though this initial attempt was followed by a 
Christmas Day attack on Chitagong by about fifty aircraft, nothing more was 
done, and again the Japanese laid down a weapon to which northeast India and 
the Assam line of communications were most vulnerable. After the war, the 
Japanese explained that though they had fully appreciated the importance of 
bombing Calcutta, they were handicapped by the lack of bombers. Of the four 
Japanese bomber units in Burma, two were transferred to New Guinea in late 
January 1944, leaving only fifteen light and nine heavy bombers. With the 
bomber units went three fighter regiments. What remained was not enough 
for sustained and effective operations. 9 

The interval before preparations began to stage for the proposed Japanese 
offensive against India was used by the Japanese airmen for attacks on the 
Hump airfields and on the transports themselves. Tinsukia airfield was at- 
tacked by about fifty fighters and eighteen bombers on 8 December, and Kun- 
ming on the 18th and 22d of December. Beginning mid-January 1944, Japa- 
nese pilots made an effort to attack ATC transports as they were spotted by 
Japanese observers in and around Sumprabum in north Burma. The effort 
seemed "a serious threat" to EAC, but it was short-lived because in May the 
Japanese lost the Myitkyina airstrip from which the attacks had been made. 10 

When the third phase of the Japanese offensive began with a diversionary 
attack in the Arakan, the 12th, 64th, and 204th Air Regiments co-operated by 
making sweeps over the battlefields. 11 Where classic air doctrine called for at- 
tacks on the hostile airfields and forces until air superiority had been attained, 
the Japanese interpreted air-ground co-operation in the most literal sense by 
ignoring the airfields in order to engage in fighter sweeps over the battlefields. 
When the sound of the engines died away, the battle resumed as though the 
Oscars had never passed above. "The period in which we [Japanese] held con- 
trol of the air [the EAC] avoided us and after we had returned to our base 
would appear in the air over the battlefields. As a result, they disrupted the 
attacks of our ground army. Taking advantage of the unfavorable conditions 
existing when we attacked the ground with our fighters, they would challenge 

8 (1) Japanese Study 94. (2) EAC Despatch, p. 11. (3) Japanese Officers' Comments, p. 8. 

9 (1) Bull cited ! n. M 471 (2) USSBS, Japan ese Air Po wer (Washington, 1946), p. 20. (3) 
Japanese Study 94. {TyTftfwell's Mission to China]^^^^Q( 5 ) Japanese Officers' Comments, p. 9. 

10 (1) Japanese Study 94. (2) EAC Despatch, p. 17. (3) Japanese Officers' Comments, p. 10. 

11 It will be recalled that the first phase was the elimination of the Chinese bridgehead across 
the Salween River in October 1943 and the second phase was the attempt by the 18th Division to 
occupy the upper Hukawng Valley. The purpose of these moves was to secure the fknjyjyjdrear 
of the Japanese forces making the main effort. See Stilwell's Mission to China, pp. 1 353-54. | 



us to an air battle. Thus, with the passing of time, battle became more diffi- 
cult." 12 For February 1944, EAC claimed only 15 Japanese aircraft certainly 
destroyed and 6 probably so, as against 25 of its own lost, but the Japanese at- 
tempt was a failure for day after day the transport aircraft flew in supplies for 
the divisions trapped by the Japanese thrust. 13 And, when the British com- 
manders had stopped the Japanese offensive in the Arakan, they were able to 
disengage the 5th and 7th Indian Divisions from the Arakan and rush them 
north by rail and air to meet the next phase in the Japanese offensive. Thus, air 
transport gave Generals Giffard and Slim an advantage in strategic mobility 
which the Japanese could not match, and which the two British commanders 
exploited to the utmost. 14 

The 5th Air Division's support of the main Japanese thrust on Imphal was 
more skillfully handled. On 12, 15, 16, and 17 March, the Japanese attacked 
the airfields from which EAC was supporting 4 Corps on the Imphal front. 
They bombed the airstrips used for the flying in of an ambitious attempt by 
Maj. Gen. Orde C. Wingate's Long-Range Penetration Groups on 10, 13, 16, 
17, and 18 March. These operations were mixed with attacks on fields in the 
Arakan, at Tinsukia, and at Ledo. March cost EAC fifty-seven aircraft destroyed 
and missing. Eastern Air Command headquarters believed the Japanese lost 
fifty-nine in the air and fifty-eight on the ground. The Japanese account is in- 
complete but suggests that EAC claims are exaggerated. 

However, one Japanese operation was an utter fiasco and minor disaster. 
The 62d Air Regiment, nine heavy bombers, was brought to Burma to attack the 
Ledo area. With an escort of about sixty fighters, the 62d Air Regiment made its 
attempt on 27 March 1944. Lacking radar bombsights, the Japanese loosed their 
bombs at random through a heavy overcast. EAC fighters attacked the force on 
its way home when it was low on fuel. Eight of the bombers were shot down 
and the ninth made a forced landing. 

Though the Japanese at the end of March had failed to prevent EAC from 
maintaining whole divisions by air supply, a feat impossible against a foe that 
has air superiority, they were still very much in the fight against an Allied op- 
ponent that, during March, had a daily operational average of 582 fighters. The 
Japanese aircraft replacement system worked well from January to June 1944, 
giving 3d Air Army from 100 to 150 aircraft a month, most of which went to 
Burma. The trained aircrews were not replaced. 15 

If the Japanese in March did feel a certain satisfaction at having held their 
own against very heavy odds, their complacence must have been short-lived, 
for in March EAC was only beginning to hit its stride. After March, the bal- 
ance tilted in its favor ever more rapidly. Eastern Air Command dispatched 
8,353 sorties in January 1944 and 18,109 in March. By May the figure was up 

12 Japanese Study 94, p. 58. 
1} EAC Despatch, p. 18. 

14 Mountbatten Report, Pt. B, pars. 99, 106. ^^^^^ 
" (1) Japanese Study 94. (2) EAC Despatch, App. 1.(3) Bull cite jn. <>(*>)~\ 



to 23,490. EAC dropped a modest 1,513.4 tons of bombs in January 1944, 3,189 
tons in March and 4,073.7 tons in May. American components of EAC ex- 
pended 279,000 rounds of 50-caliber ammunition in January, 675,000 in March, 
1,280,000 in April, and 867,989 in May. The RAF was proportionately active. 

This steady acceleration of effort, reflecting among other things an increase 
in the AAF components of the EAC from 297 aircraft in April to 444 in May, 
weighed ever more heavily on the Japanese. After June their losses in fighters 
forced them to cease attempts to intercept EAC's attacks, an admission of de- 
feat. After the war, the Japanese considered that "well-planned aerial combat 
tactics and superior planes always gave the enemy an upper hand in the engage- 
ments." Japanese sorties in support of the attack on Imphal, and against Win- 
gate's Chindits, which had reached a peak of 570 sorties in April, fell to 423 
in May, and plummeted to 115 in June. 16 

The drop in Japanese sorties was much sharper than that which the mon- 
soon forced on EAC. Whatever the exact count of sorties made by the Jap- 
anese, their effort during the crucial month of their attack on India was but a 
fraction of EAC's, and there was not a chance of their attaining the air supe- 
riority they held so often in 1941 and 1942. Indeed, EAC controlled the air 
space over the battlefield, which meant that the Japanese could no longer 
match the Allied ground forces in strategic mobility, nor was the most impor- 
tant Japanese tactic in jungle warfare, encirclement, the menace it had once 
been. Now, thanks to EAC's air superiority, the Allies could fly whole divi- 
sions to the scene of action, while the Japanese struggled to bring troops for- 
ward over bomb-damaged rail lines or on foot over muddy trails. On the battle- 
field itself, the Japanese could surround Allied units still, for they had all their 
old skill and aggressiveness in maneuver, but the success was an empty one, for 
the air transports could always deliver supplies to the surrounded unit. This 
was a grim omen, for when the Allies were on the offensive, air supply meant 
that Chinese, Indians, British, and Americans could operate independently of 
ground lines of communications, could move freely past the Japanese flanks 
into the Japanese rear areas, and there block the roads that fed the Japanese. 
When this happened the Japanese would have to cut their way out or starve, 
for EAC's fighters and bombers would see to it that no Japanese transports ap- 
peared over the battlefield. 

Support of Ground Forces 

In the difficult terrain of Burma, where artillery was hard to bring forward 
and tended to lessen the mobility of Allied units, EAC's fighters, fighter bomb- 

16 (1) EAC Despatch, Apps. 3-7. (2) Japanese Study 94, p. 67, Table, pp. 62-65. Japanese 
statistics give the number of aircraft dispatched by day. The 8th, 50th, 64th, and 204th Air Regi- 
ments supported the Japanese in north Burma, with about fifty-five fighters at the beginning of 
March 1944, dropping to about thirty-two planes at the end. The table suggests each fighter aver- 
aged 4.2 sorties. That month EAC fighters averaged 21, over the whole of Burma. Even if the 
Japanese fighters are credited with sorties along the Salween, and over Myitkyina and Ledo, 10 
sorties a month would be a generous estimate, for in none of these other areas was 5th Air Division 
an active factor. Very probably the Japanese were greatly handicapped by maintenance problems. 



ers, and bombers of all types acted as substitutes. Suitable targets, according to 
current local practice, were troop concentrations, bunkers, machine gun nests, 
artillery, towns and villages sheltering the Japanese, and supply convoys. The 
choice among types of bombs and ammunition was governed by the nature of 
the target. Where possible, cover was used in the approach, and attacks were 
made from medium or low altitude. Despite the seemingly obvious worth of 
close tactical air support, it was not common or particularly effective before the 
spring of 1944. Earlier, the priority demands for bombing transport, factories, 
and air installations had absorbed most air resources. There had also been the 
belief that fighter aircraft should confine themselves to missions of escort and 
interception. Last of all, there had been no effective mechanism whereby the 
ground forces could obtain air support on call. 

By May 1944, Northern Sector Air Force, the AAF component of the Third 
Tactical Air Force, had worked out a technique for air support in siege or mo- 
bile warfare. For the former, A- 2 and A-3 were responsible for joint planning 
with the task force G-2 and G-3. The latter two chose the targets, while the 
air staff "planned the attack, determined the number and type of aircraft to be 
used, the types of bombs, the techniques of attack, the selection of the units 
. . . and the briefing of the crews." For mobile warfare, Northern Sector Air 
Force provided the "air party": 

Personnel of an air party consisted of a team of one or two officers with six to eight en- 
listed men. Their station was with an advanced brigade at the front. . . . They were in 
[radio communication] with the air office, giving immediate information on targets selected 
by the army and approved by the air party. They also served as guides to aircraft which were 
making the air strikes. Sometimes, when it was impossible to observe the tactical situation 
from the position of the air party, an L-5 was used for strike observation, which worked 
through the air party. 17 

Radio and photography were vital in the air-ground support mechanism. Of 
the several air elements in Burma, the Tenth Air Force reached the highest 
point in air-ground radio communication, using high-frequency SCR-117 or 
-118 for point-to-point transmission linking the air party with the air force's 
signal center, and very high frequency radio for communication with the air- 
craft making the attack. "In cases of emergency, it took about twenty minutes 
for communications to be sent from the air party to fighter group headquarters. 
In some cases aircraft were over the target thirty minutes after the original re- 
quest." The Tenth Air Force also made complete use of photographic facilities. 
Low-level verticals, reconnaissance strips, obliques, and pinpoint shots were 
used. Both air and ground personnel used photographs for target designation, 
briefing, and aerial identification of targets. 

Of equal importance in target designation were the efforts of the Kachins, 
Burmese, and Americans working for the Office of Strategic Services in De- 
tachment 101. Operating behind the Japanese lines, and well equipped with 

17 EAC Despatch, pp. 72, 79-80, 81; quotation on p. 82. 



radios, these daring men sent a steady flow of data on the location of supply 
dumps, rest areas, billets, command posts, and the like. 18 

About one third of the fighter sorties were in direct support of the infantry. 
This meant 221 Group flew 9,000 such sorties between March and July 1944, 
while their colleagues of 224 Group flew about 5 ,000. During siege operations 
in north Burma, Northern Sector Air Force and successor commands averaged 
thirty-three fighter sorties a day, a rather deceptive figure, for these sorties were 
flown in monsoon weather so that good flying days saw many times the thirty- 
three figure flown. The RAF expended 25,000 rounds of40-mm. ammunition 
in ground support during the first six months of 1944. In the same period the 
fighters dropped 9,327 tons of bombs. 19 

Air-ground co-operation was most marked in its effect on the Japanese 13 th 
Army's attack on India in spring 1944. To obtain surprise, the Japanese delib- 
erately omitted much necessary preliminary engineer work on the dirt trails 
from the Chindwin River forward that were to sustain their three divisions in 
battle. The Japanese planned on a grand rush that would carry them over the 
British supply dumps in a matter of weeks. The opening phases went well; 
then the battle settled down to hard pounding. At this point the 4 Corps at 
Imphal depended on air supply, 33 Corps driving to relieve it depended on the 
Bengal and Assam Railway and Dimapur road, and 13th Army on the jungle 

Allied fighters swarmed over the dirt trails all day long, bombing and 
strafing. Japanese supplies could be moved only at night, a task fantastically 
complicated when the monsoon rains began. As a result of the constriction of 
the roads, from May 1944 on 13th Army quite literally starved. Looking back on 
their experiences at Imphal, the Japanese commanders and staffs approached 
eloquence in dilating on the importance of air-ground co-operation: "With a 
good signals system and air supplies, the Allies were able to carry out their 
operations freely and unhindered whereas the Japanese without air supplies and 
with their only means of supply— ground transport— cut off, were in a para- 
lyzed state. . . . The difference in ground-air cooperation between the Japanese 
Army and the Allies was the difference between victory and defeat." To this 
comment by Lt. Col. Iwaichi Fujiwara, sometime G-2 and G-3 of 13th Army, 
could be added similar ones by Lt. Gen. Takazo Numata, Chief of Staff, 
Headquarters, Southern Army. 20 

A form of air support most highly valued by the infantryman was the air 
evacuation of ground casualties on the grand scale. Where previously the 
wounded, in default of other means of transport, had jolted and swayed for 
agonizing days in litters, ambulances, and trains on the long trip back to hos- 

18 (1) For a more detailed account of Detachment 101, s e j Chapter IJ above. (2) NCAC 
History, App. 8, Brief Sketch of Detachment 101 in the NCAC Campaign. 

19 EAC Despatch, pp. 84-86, 87-88; Apps. 5, 6. 

2n Col Fujiwara, SEATIC Bull 240, 9 Jul 46, pp. 23-24, 51; SEATIC Bull 242, 7 Aug 46, pp. 
31-32; SEATIC Bull 247, 22 Apr 47, p. 25. MID Library. 



pital, now small liaison aircraft could land on a rough airstrip immediately 
behind the lines, pick up the soldier, and fly him to a collecting point. There 
an air transport with a flight nurse would receive him and fly him back to hos- 
pital. The worst part of the journey would be that to the first airstrip for if the 
soldier was wounded in the hills it might take the litter bearers several days to 
cover the seven or eight miles to the nearest airstrip. If the soldier in north 
Burma happened to be wounded near the combat supply road which ran south 
and east through the Hukawng and Mogaung Valleys, he could be taken by 
ambulance to the clearing company, and by ambulance again to the American 
field hospital. Evacuation of British and Indian units followed a similar pattern; 
the same facilities served all Allied troops. Between December 1943 and August 
1944, 18,256 casualties were evacuated from north Burma by air. From Burma 
as a whole, EAC in sixteen months evacuated more than 106,000 casualties. 21 

Strategic Bombing 

As noted before in this history, Burma is not an ideal subject for the 
demonstration of strategic bombardment, for it is very large, and predominantly 
agricultural. The Japanese Army depended on the Burmese economy for food, 
oil, some clothing, and a few simple munitions and spare parts. After October 
1943 and the opening of the Burma-Siam Railway almost everything else came 
in by train. Of the targets found in Burma, the oil fields, the supply dumps, and 
the rail lines seemed the most promising. The Japanese Army depended on its 
infantry, so its oil needs were modest, and even sharp reductions in the capacity 
of the Yenangyaung oil installations seem not to have hurt Burma Area Army's 
supply position. Japanese dumps closer to the front suffered heavily in the all- 
important Manipur area, but not elsewhere in spring 1944. The railroads, to 
Allied target analysts, seemed clearly indicated as the most promising target, 
though not an easy or fragile one. Against these several classes of targets for 
strategic bombing, Strategic Air Force in January 1944 could pit 61 B-24's, 
25 Wellingtons, and 32 B-25's. In June, the command had 53 B-24's, 13 
Wellingtons, and 69 B-25's. Such was not an imposing force with which to 
attack an area the size of Germany. 22 

"The main purpose of the Strategic Air Force . . . was to disrupt the 
enemy's entire transportation system in the India-Burma Theater. . . ." The 
initial EAC directives gave first priority to hostile air force installations, but in 
practice this mission fell to the Tactical Air Force. Shipping was given second 
priority, for its highly mobile nature made it a target of opportunity. The rail- 
roads were always present, and they had third priority. After the opening of the 

21 (1) EAC Despatch, p. 90. (2) NCAC Air Supply, Vol. I, App. 4. 

22 EAC Despatch, App. 2. These are daily operational averages. The RAF seems to have had 
maintenance problems. Rarely was more than 50 percent of its contribution operational. 



Burma-Siam Railway Japanese shipping disappeared from the Rangoon area 
and sea sweeps were not profitable. As of 31 March 1944 the 7th Bombardment 
Group (H) had dropped only nineteen tons of bombs on shipping in the 
Rangoon area. In practice, then, until June 1944 the railroads were the number 
one target for Strategic Air Force. Then Operational Directive 10 gave first 
priority to support of Fourteenth Army (the British formation on the Manipur 
front) and second to attacks on shipping and railways. 23 

The Fourteenth Air Force was not an appreciable factor in isolating Burma 
through destruction of Japanese shipping. On the other hand, its operations 
did introduce the element of delay and uncertainty into the southward move- 
ment of Japanese supplies by sea. Eastern Air Command's antishipping opera- 
tions by bomb and mine made the Japanese happy to leave the task of supply- 
ing the Burma Area Army to the Burma-Siam Railway. Up to spring 1944 the 
Japanese maintained a limited traffic in small wooden ships between Singapore 
and Rangoon. Then the increasing efficiency of the Allied bombers restricted 
them to night operations along the coast line, and finally toward the war's end 
reduced them to using luggers and farmers' skiffs ("country boats") between 
Moulmein and Rangoon. 24 

In attacking land communications, Strategic Air Force followed four 
principles: (1) to attack railways as the most important such target within reach, 
(2) to pick bridges as the most vulnerable point on the railway, (3) to isolate 
segments of line and destroy rolling stock and engines trapped therein, and 
(4) to use diversity of attack to confuse the enemy. 

Bombing the railways did not bring major results in the 1943-44 campaign 
because the means at the airmen's disposal were inadequate, though they 
attacked with devotion and skill. On New Year's Day, 1944, B-25's blew out 
140 feet of the 400-foot Mu River span, vital for carrying supplies toward the 
Chindwin area. The weather turned bad in January and 50 percent of the 
missions were canceled. The 7th Bombardment Group was withdrawn for 
special training in February. Its return in March, plus the addition of the 12th 
Bombardment Group, made for a greater weight of attack. The 1,650-foot 
Sittang Bridge, site of the 1942 tragedy, was crippled in April 1944 as soon as 
the Japanese repaired it after months of effort. The 700-foot bridge just north 
of Mandalay on the line to Myitkyina was successfully attacked. With these 
bridges went more than a score of others. When the Japanese moved against 
India, the bombers switched their attention to the roads west from Ye-u and 
Wuntho and pounded them daily. Attacks were also made on the Yenangyaung 
oil fields. 25 

The 7,348.3 tons of bombs dropped by the Strategic Air Force up to 30 June 

23 (1) Activities of the 7th Bomb Gp (H), Tenth Air Force, 1 Feb 43-31 Mar 44, MS, p. 1. 
KCRC. (2) EAC Despatch, pp. 47-48. 

24 (1) For an analysis of Fourteenth Air Force claims see Stilwell's Mission to China X Chart 8. I 
(2) Japanese Study 116. (3) SEATIC Bull 247, p. 28. MID Library. (4) EAC Despatch, pp. 53-58. 

25 EAC Despatch, pp. 58, 60-63- 



BOMBING OF BRIDGES such as this Mu River bypass bridge slowed the movement 
of Japanese supplies to the Chindwin area. 

19AA did not affect the campaign because they did not disrupt Japanese rail 
traffic. Makeshift repairs and gangs of coolies could always carry supplies over 
breaks in the lines. The four main sectors on which the Japanese fought in 
Burma shared one common peculiarity in that each was separated from the 
railhead by many miles of unimproved road. In the case of the 18th Division in 
the Hukawng Valley 140 miles lay between it and the railway. The 13th Army 
left its railhead even farther behind when it attacked toward Imphal, and 
further complicated its supply problems by crossing a major river and a moun- 
tain range. When 13th Army failed to take the British supply dumps in the first 
rush it simply thrust its head into a noose. The 28th Army in the Arakan 
depended on a combination of dirt trails and coastal waterways. Only the 36th 
Division on the Salween front had a good supply situation and it was signifi- 
cantly near the Lashio railhead. Until the 3d Indian Division was dropped 
astride the railway to Myitkyina that line had no trouble in bringing supplies 
to the 18th Division's railhead. 

The ultimately crippling Japanese supply problems on the Arakan, Manipur, 
and north Burma fronts began at railhead, with the inherent limitations of a 
dirt road many times compounded by a steadily worsening truck shortage, 



Table 1 — Tonnages Shipped to Burma Over Burma-Siam Railway: 
November 1943-August 1945 

Year and month 














Short tons 

228, 550 

9, 500 

12, 350 
10, 150 
15, 650 
17, 100 

Year and month 














Source: SEATIC Bull 246, Burma-Siam Ry, 3 Oct 45. MID Library. 

monsoon rains, and the constant attacks of Allied fighters and bombers. 26 
Japan's principal supply route to Burma, the Burma-Siam Railway, was not 
affected by air bombing in the first six months of 1944. Traffic over it increased 
steadily. Allied prisoners of war later reported that on occasion they saw as 
many as ten trains a day moving in each direction. After the war, the Japanese 
claimed that traffic on the Burma-Siam Railway increased from a daily average 
of one train in December 1943 to three trains in June 1944. 27 (Table 1 ) 

The coming of the monsoon rains and the diversion of some of the U.S. 
B-24's to air cargo operations brought a sharp reduction in strategic bombing. 
Tonnage dropped from the 2,069 mark of May to 285 in July and 206 in 

26 In the peak year, 1942, Japanese truck production was but 35,386 units. The shortage of 
trucks was general in the Japanese Army. At the end of the war, the Army in the home islands had 
but 8,900 trucks. USSBS, The Effects of Strategic Bombing on Japan's War Economy, pp. 65, 220. App. 
Table C-l 58. 

27 Three sources were drawn on for the above appraisal. (1) Interrogations of Japanese senior 
officers conducted immediately after the war by SEATIC are in Bulletins 246 and 247, Military 
Intelligence Division Library. Among the officers questioned were: Lt. Gen. Masaki Honda, Com- 
manding General, 33d Army; Lt. Col. Masayuki Taguchi, 56th Division staff; and General Tanaka, 
Commanding General, 18th Division. (2) Questionnaires, prepared by the authors, were answered 
by the above Japanese officers in 1948, plus Lt. Gen. Kunomura Momoyo, 15th Army staff; Lt. 
Gen. Eitaro Naka, Chief of Staff, Burma Area Army; Lt. Gen. Tadashi Katakura, Chief of Staff, 
33d Army; Col. Fujio Kawamichi, Chief of Staff, 56th Division; Maj. Iwao Takahashi, 15th Army 
staff; and Lt. Col. Shinroku Iwakoshi, Chief, Imperial Headquarters Supply Section. All replies are 
in OCMH Files. (3) SEATIC Bull 246 has a wealth of information from prisoner-of-war and 
Japanese sources. 



August, rising thereafter to 4l6 in September. Tonnages did not hit the 1,000 
mark for the rest of the year. 28 

Because the U.S. combat effort in Burma was primarily in the air, an 
appraisal of air operations cannot be omitted in a theater history. Not until the 
fall of 1944 were there two U.S. regimental combat teams in Burma, while in 
the air the United States contributed two air forces, the B-29's, and the 
tremendous Air Transport Command effort. It is notable that an enormous 
disparity of strength in the air could not prevent the Japanese from advancing 
at will in the Arakan and Manipur sectors until they reached the point at 
which the Indians and British planned to hold and fight it out. 

When the battle on the ground was fairly joined, air supply, which is 
possible only if the air space over the battlefield is controlled by friendly hands, 
prevented the Japanese encircling tactic from defeating the Allies once more. 
Allied control of the air permitted continual attacks on the roads supplying the 
Japanese, thus making their supply situation on the Imphal front ultimately 
impossible. In north Burma, the encircling tactics that Stilwell used against the 
Japanese forced them, in the absence of their own air supply, to fight a series 
of retreating actions with heavy losses in equipment and supplies, further 
complicated by the deficiencies of their primitive line of communications. 

It is noteworthy that the heaviest Japanese rail traffic coincided with the 
heaviest EAC bombing effort, th en fell off s harply in October 1944 when but 
500 tons of bombs were dropped. (Table 2 ) The decline immediately followed 
the monsoon rains in which bombing had been light and sporadic. The great 
slump in Japanese rail traffic in Burma did not come until January 1945, when 
tactical aviation was able to cover the rail net as in spring 1944 it had 
interdicted the jungle trails, when Allied troops were progressively over- 
running the railways, and when the Japanese shipping situation was approach- 
ing the catastrophic. The contrast between the relative futility of the strategic 
bombing effort in the 1943-44 campaign and the success of air supply, indicates 
that air supply, made possible by air superiority, is the key to Allied victory 
in Burma. 29 

Air Supply 

The nature of Burma's terrain and climate and the Japanese military system, 
which had taken such clever advantage of them in the 1942-43 campaigns, 

28 EAC Despatch, App. 4. The B-24's carried 2,621 tons of cargo in September 1944. 

29 (1) Bull cited n. 27(3). USSBS, The Effect of Air Action on Japanese Ground Army Logistics 
(Washington, 1947), p. 57. The tonnage chart on the page cited was drawn by a Japanese officer 
in 1946. Page 32 in the text says that the decline in traffic is attributable to the monsoon's washing 
away the makeshift bridges which replaced those damaged by bombing. The chart shows in 1944 
and, significantly, 1943 a sharp drop beginning in July- August, or one or two months after the 
monsoon began. The chart shows this seasonal dip to be an annual feature, falling to the same level 
every year. Whereas traffic in 1943 recovered sharply, it plummeted in 1944, for late in that year 
large segments of the rail net came under Allied control. 


Table 2— Average Daily Traffic at Terminals of Principal Branch 
Lines of Japanese-Operated Railways in Burma: 
January 1943-August 1945 

[Number of Railroad Can] 

Yetr and tnomh 




My Sn- 



Siim • 































































































t Ci 












February ...»,„. , 

















April ....... . L . . . , . . . 








( s ) 



















July. ... T .......... . 








( 4 ) 





































December. .......... 






























April , , 









June . . t . , r 




J u| y 








• Eittmitei of Surma Arta Army, which did POt control Iriffic on Burmi-Siim line. 

* Ftfuru on i available. 

^ Sount; USSBS, Alchd %Mt 1. f*t National ArchivM. Dtily IVtraa* did Wd on r«ordi it gurme Jrt, 



impressed themselves strongly on the senior Allied ground force officers in 
India and Burma. The geographic factors of heavily wooded jungle, rough 
terrain, and monsoon climate greatly aggravated the problems of moving sup- 
plies on the ground. Roads, when made, had to be maintained, and a great deal 
of the road's capacity was consumed by maintenance, thus reducing its pay 
load. A more economical system for the supply of forward elements was 
needed. In the light of these factors, by summer 1943 the American members 
of the staff of the Chinese Army in India explicitly accepted the principle that 
when the North Burma Campaign began, troops in the forward area would be 
supplied by air.^° 

The experiments that finally led to this conclusion began with the airdrops 
made to refugees fleeing Burma in May, June, and July 1942. That summer, 
General Wingate organized his first Long-Range Penetration Group to operate 
solely on air supply. Concurrently, a small unit of Australian troops, retreating 
from the Japanese across New Guinea's Owen Stanley Mountains, was supplied 
by air. When Allied strength in New Guinea permitted a counteroffensive, the 
2d Battalion, 126th U.S. Infantry, was supplied by air on its way back over the 
mountains. Wingate's February 1943 expedition was supplied by air. 31 

With these experiences in mind, Stilwell in late February 1943 resolved that 
isolated troops in the Naga Hills, such as air- warning stations and Chinese 
outposts guarding the Ledo Road, should be supplied by air. Details came from 
laundry and ordnance personnel, the 51st Fighter Group supplied containers 
and parachutes, the Ferry Command at Chabua assigned some C-47's, and a 
junior officer, 1st Lt. Frederick L. Wood, Jr., of Ordnance, was put in charge. 

The experiment worked so well that at the end of the year permanent 
organization seemed advisable. This covered simply the warehousing, packing, 
dropping, and receipt of supplies. By September 1943, the air supply of combat 
in north Burma was accepted as a matter of course. G-4 of the Combat Troops, 
Ledo Sector, was responsible for target designation and proper delivery. SOS 
stored and packed. The air forces delivered. No one organization was charged 
with sole responsibility. This division promised future administrative 

When the Chinese Army in India opened the North Burma Campaign in 
October 1943, the air supply resources committed to the effort were: 1st and 2d 
Troop Carrier Squadrons; 518th Quartermaster Battalion (Mobile), charged 
with procuring, warehousing, packing, and dropping; the 384lst QM Truck 
Company, which did the hazardous and responsible work of kicking supplies 
from the transports on schedule; and the 3304th QM Truck Company, which 
received and distributed. These units were under Base Transportation Section, 

30 (1) Opn Plan Albacore, 8 Aug 43. Folder, Albacore, Ledo Striptease, NCAC Files, 
KCRC. (2) NCAC Air Supply, pp. 8-1 to 16-1. 

M (1) Frank Owen, The Campaign in Burma (London: His Majesty's Stationery Office, 1946), 
p 45. (2) Rpt, G-3 Section During Papuan Mpw f^iiinpa Campaign, 2 Oct 42-24 Jan 43. AG 
332-333.4. (3) Stilwell 's Mission to China jCh. VIII. | 


Base Section No. 3. The problems involved rapidly passed beyond its scope 
and in October the base section S-4 took over. 32 

Allocation of Transport Aircraft 

By the end of December 1943, air deliveries had increased from the 199 
net tons of April 1943 to 1,391 tons. In percentage the increase was great, but 
the greatest development lay in the future. As part of the integration of the air 
forces, Troop Carrier Command was activated 15 December 1943 under General 
Old, with headquarters at Comilla. Under him were the 1st and 2d Troop 
Carrier Squadrons listed above, plus the 27th and 315th Troop Carrier 
Squadrons, and 31, 62, 117, and 194 Squadrons of the RAF. 33 

Mountbatten was fully aware of the possibilities of airborne operations, 
while his headquarters was coming to appreciate what might be done by stand- 

32 NCAC Air Supply, pp. 3 5— III, 5— III, 7— III, 10— III. Though the kickers shared every hazard 
of the aircrew, they long performed their dangerous duties without Air Medals, Distinguished 
Flying Crosses, flight pay, a fixed number of missions, or any of the recognition and perquisites 
given Air Forces personnel. Many kickers had more than 1,400 flying hours, while a considerable 
number had more than 1,000 hours. 

» NCAC Air Supply, p. 1 l-III. 



ard infantry divisions on air supply. He sought energetically to increase the 
allotment of transports for SEAC and saw the ATC as a source of aircraft for 
the proposal to drop a division on the rail line to Myitkyina. The initial agree- 
ment was that Mountbatten could divert the equivalent of 1,100 tons a month 
from the ATC to furnish air supply to his campaign. Later, the Joint Chiefs of 
Staff suggested that while the CCS should recognize that SEAC must have 
resources adequate to its tasks it was also necessary to have a firm commitment 
to China over the next six months for ATC tonnage. Specifically, the JCS pro- 
posed that SEAC's transport allotment be increased from 11 C-47's or 8 C-46's 
in February 1944 to 188 C-47's or 126 C-46's in the following month, when 
major offensive operations were to take place. The CCS approved, but since 
the expected major operations vanished when the Generalissimo refused to 
cross the Salween, there were only seventy-six transports on hand for SEAC 
when March began. 34 

Mountbatten did not regard seventy-six planes as adequate, and his repre- 
sentatives in "Washington were asked to press for more, on the grounds that his 
resources would not permit him to take and hold the Myitkyina area as the 
JCS desired. He told them: 

After seeing the performance of Stilwell's Chinese forces and hearing of the wonderful 
show which Wingate and Cochran's No. 1 Air Commando have put up I am becoming 
convinced that Allied Forces could march all over Burma provided they have adequate air 
supply and air support and I hope that you will do all you can to emphasize the need for 
more transport aircraft and yet more transport aircraft and the rapid formation of further Air 
Commandos on Cochran's style both in England and America. 35 

To Mountbatten's request, thus relayed, General Arnold replied that he was 
forming four transport groups totaling 400 aircraft. The first 100 planes went 
to SEAC in May 1944. Thus SEAC began to acquire the great fleets of air trans- 
ports which so changed the nature of war in Burma. The difficulty in the way 
of Mountbatten's having the full complement that he requested lay in his 
omitting to support his request with detailed plans for future aggressive opera- 
tions. The Joint Planners, in default of such data, preferred to recommend that 
SEAC have but 200 transports and Gen. Douglas MacArthur in the Pacific, the 
balance. 36 

As for the type of transport aircraft used, the C-47 had qualities of stability 
that resulted in its being assigned almost all airdropping missions. It was easy 
to load and balance the C-47 so long as cargo was tied down as far forward as 
possible. When the aircraft was in flight, cargo could be dragged to the door 

34 (1) CCS 411/1, 26 Nov 43, sub: Opns in SEAC. (2) Rad SEACOS 89245, 1 Dec 43. ABC 
384 (Burma) 8-25-42, Sec IV, A48-224. (3) Min, CCS 138th Mtg, 7 Dec 43, Item 8. (4) CCS 
411/5, 7 Dec 43, sub: Opns in SEAC. (5) Rad RE-240, Brig Gen Vernon Evans, DCofS, USAF 
CBI, to Stilwell, 1 1 Feb 44. Bk 6A, JWS Personal File. 

" (1) Mountbatten Report, Pt. B, p. 73, par. 138. (2) Quotation from Minute SC4/440/W, 
SAC for Wedemeyer, 16 Mar 44. SEAC War Diary. 

36 (1) Min, CCS 151st Mtg, 24 Mar 44, Item 3. (2) JCS Memo for Info 302, 11 Sep 44, Incl 
C, par. 5. 



and kicked out without upsetting the transport's balance. In the case of the 
C-46, cargo had to be arranged in the fuselage literally with slide rule precision 
lest the craft's flying qualities be adversely affected. Nor could cargo be shifted 
readily when the aircraft was in flight, for the same reason. Therefore the C-46 
and the four-engined C-54 were most useful on the Hump, while the stable, 
sturdy, and dependable C-47 was excellently adapted to dropping supplies from 
low altitudes. 37 

Administrative Problems 

In north Burma, the SOS continually altered and enlarged its organization 
to keep pace with the expanding needs of the combat troops for air supply. 
Base Section No. 3, under Colonel Pick, recognized that air supply had out- 
grown the stage at which a quartermaster battalion could meet the responsi- 
bility, and decided to activate a packing and airdropping unit on an ad hoc basis. 
The 3962d and 3964th QM Truck Companies, of the 518th QM Battalion 
(Mobile), plus kicking crews from the 384lst QM Truck Company began 
operations on 1 December 1943 from the Sookerating Tea Factory, Sookerating, 
Assam, near the airfield of that nam e|f5Vg Map It is worthy of note that all 
air-delivery equipment used in the CB1 Theater was of British or Indian origin. 
Not until Galahad's air-supply personnel arrived at Dinjan airfield in late 
January 1944 did the air-supply personnel have a chance to examine U.S. -made 
airdrop equipment. 38 At the front, matters were simplified by a decision to drop 
directly to the troops, relieving the quartermaster personnel drawn from the 
3304th QM Truck Company of their mission of receiving and distributing. In 
February 1944, Base Section No. 3 set up an airdropping section under its S^4 
to perform its share of the administrative work of air supply. 

Within EAC, the mechanism initially set up by Troop Carrier Command to 
apportion aircraft among the several fronts did not work well. Aircraft were 
allocated on the basis of advance estimates of a month's requirements in terms 
of air-supply tonnage. In practice, it proved impossible for any commander to 
estimate the urgency of his needs fifteen to forty-five days in advance. The 
result was an inflexibility wasteful of aircraft. G-4 of the Sino- American combat 
troops in north Burma had to be forever appealing to Troop Carrier Command 
to alter its schedules on the basis of a changed tactical situation, while 
simultaneously informing SOS of the degree of priority attached to each 
mission. 39 

The rigidity of this system of aircraft allocation was no doubt a partial cause 
of the shortage of aircraft supporting the North Burma Campaign. Even the 

37 Notes of Joseph Bykofsky, Transportation Section, OCMH, on draft manuscript of this chap- 
ter. Mr. Bykofsky was in an air supply organization during the later phases of the North Burma 

38 NCAC Air Supply, pp. 7-I1I to 10-111. 

39 NCAC Air Supply, pp. 6-III to 13-111, 17-111 to 20-111. 



opening of a subdepot at Shingbwiyang airstrip in the Hukawng Valley, 
January 1944, though it shortened turnaround time, did not relieve the 
problem. General Boatner requested Base Section No. 3 to build the supply 
levels in the forward area up to the required point. That his request, with its 
implied demand on aircraft, went to SOS, implies the threefold nature of the 
air-supply mechanism at that time. A fortnight later Boatner wrote to theater 
headquarters, pointing out that he had 14,000 men in advance of the roadhead 
and all dependent on air supply, and that a regiment, a tank group, and a 
British brigade had been moved into his area with no increase in his air 
resources. This recourse yielded nothing, and so finally Boatner wrote directly 
to Stilwell. 

The theater commander promptly interceded with Stratemeyer and Old. On 
1 1 February Stilwell told Stratemeyer that the next few weeks would be critical, 
after which, progress on the Ledo Road would ease the supply situation. 
Stilwell understood that troop-carrier aircraft had been diverted, tonnage into 
Fort Hertz increased without his consent, and aircraft diverted from air-supply 
missions for secret tasks without prior notice. Over a six-day period his head- 
quarters had asked for 158 planeloads and received only 98. Not intending to 
be deprived of his resources, Stilwell was depending on Stratemeyer to see that 
a failure in air supply did not interfere with his operations. 40 

The experience with this transport shortage, and the obvious effort to find 
a responsible agency, were probably among the reasons for the decision to 
reconstitute the Tenth Air Force as the organization supporting StilwelPs cam- 
paign. No criticism of the air forces was implied in Boatner's letters, for it was 
recognized that the energy, skill, and valor of the pilots and crews in delivering 
supplies through all sorts of weather, and meeting the demands of units that 
were at times improvident in their use of air supply, often made good the errors 
of the other agencies. It was an experimental period. 

From January to May 1944, the combat headquarters in north Burma, the 
SOS, and Troop Carrier Command argued among themselves over the admin- 
istration of air supply to Stilwell's forces. Each had good reason for its claim 
to be the co-ordinating agency. SOS could contend that supply of the troops 
was its normal mission. The airmen considered that their possession of the 
means made them logically responsible for delivery at the right time and place, 
that SOS could well surrender the related ground operations. Stilwell's field 
headquarters was not so active on its own behalf for it did not desire to be 
responsible for SOS and AAF activities but it did insist on the needs of the 
ultimate consumers. The solution finally settled on in May 1944 was that G-4 
of Stilwell's field headquarters, Chih Hui Pu (later Northern Combat Area 
Command) should institute and operate a system of priorities. 41 

40 (1) NCAC Air Supply, pp. 13-111 to 18-111. (2) File 97-1597, 11 Feb 44. Folder, Chinese 
38th Div Rads-In, NCAC Files, KCRC. 

41 NCAC Air Supply, pp. 17-111 to 20-111. 



Matched with the reactivation of Tenth Air Force, 20 June 1944, as Strate- 
meyer's Tactical Air Force supporting the North Burma Campaign, the decision 
to give Stilwell's G-4 control over air-supply priorities meant that Stilwell 
controlled his tactical air support and was free to use his air-ground team as 
the needs of the campaign might suggest. 42 There was the utmost flexibility in 
the allocation of aircraft to meet G-4's priorities, and the requirements of the 
ground forces in general. Thus, the 3d Combat Cargo Group, AAF, which 
arrived in India 30 May, was attached to the Tactical Air Force by the order of 
20 June 1944, but supported both the 4 Corps at Imphal and the Chinese 
troops in north Burma with impartiality. The group's arrival lifted the total of 
AAF transports on hand from 112 to 191. It was an experimental unit with 150 
aircrews, some of them veteran ATC personnel. 43 


The techniques worked out by trial and error began to yield remarkable 
results by July 1944. Matters did not of course always run smoothly, either 
before or after the priority system was set up. On 7 June 1944, the American 
liaison officer with the Chinese 38th Division reported that unit had actually 
received only twenty-five supply drops of the sixty that G-4 claimed were 
delivered. The liaison officer was beginning to be annoyed. Willing though he 
was to go hungry, he was profoundly irritated at being assured supplies had 
been delivered when he was exceedingly well aware they had not. In Novem- 
ber, his successor radioed that if the complete failure of the Chinese medical 
supply system was his fault to let him know, but if, in fact, there was no medi- 
cal supply system, would G-4 please create one. The soldiers at the front were 
using parachutes for dressing wounds, and there were practically no drugs left 
in the former hospital which was now but a rest camp. 44 But if the machine of 
air supply was not foolproof, it did ultimately feed, clothe, and arm five 
Chinese divisions, one British division, an American brigade, and a host of 
service troops. 

G-4 of Stilwell's field headquarters changed its former Air Dropping 
Section to an Air Control Section to inaugurate the priorities system in late 
June. Ninety percent of all requests from units in the field were marked 
"urgent" by the submitting unit; they were of course reviewed and reclassified 
as "urgent," "first priority," or "second priority," by the airstrip or target 
designation of the point to which delivery was requested, and by the air base 
whence the load was to originate. 

The priorities sheet was teletyped early at night to all interested agencies 

42 EAC Despatch, App. 17a. 

43 EAC Despatch, pp. 123-24. 

44 (1) Rad, Col Thomas F. Van Natta, III, U.S. Ln Off, 38th Div, to Cannon, 7 Jun 44. MTF 
Msg Bk, NCAC Files, KCRC. (2) Rad 260-3155, 9 Nov 44. Folder, Chinese 38th Div Rads-In, 
NCAC Files, KCRC. 



and was used as the daily operational orders on the basis of which SOS assem- 
bled and packed the loads, and the Air Forces delivered. G-4 kept complete 
data on the supply status of all units supported, so as to equate it with the 
progress of operations as reported by G-3. The priorities once set were then 
followed up by G-4 to see that deliveries were made or to investigate failures 
if any occurred. The priorities sheet in turn permitted operational economies. 
Aircraft were loaded at night to be ready for a dawn take-off. Knowing what 
loads were available at each field made it possible for aircraft that finished one 
mission to go to a near-by field, pick up another load, and make another 
delivery before returning to the base. Operation of the priority system brought 
an immediate 50 percent increase in tonnage delivered without need for added 

SOS through study and experiment steadily improved its own methods of 
loading aircraft and packing supplies. Generally, they reflected the adoption of 
American industrial techniques familiar to many Army personnel through their 
civilian experiences. Standardized loads, production line techniques in packag- 
ing, prepacking of loads, plant layout designed to speed the flow of material, 
simultaneous loading, fueling, and crew briefing, round-the-clock operation of 
facilities, all were adopted. Suggestions for improvement in this application of 
advanced industrial techniques to jungle warfare came from all ranks, a full 50 
percent of whom were Negro troops. In emergencies, air-supply personnel 
worked twenty-four hours a day without sleep to get the loads out on time. 

There were three methods of delivering supplies— by landing the cargo 
aircraft at a forward strip, by parachuting supplies through the door of the 
aircraft, and by "free-dropping," as the me thod of dropping supplies packed 
simply in sacks came to be called. (Table 3 ) The simplest and most economical 
procedure, given the existence of a safe and suitably located airstrip, was to 
deliver supplies direct from the transport. When that method was possible, its 
economy and convenience made it preferable. When airstrips were not avail- 
able or practical, then free-dropping and parachuting were used, to which the 
items most in demand— rations, forage, fuel, and ammunition— were well 
adapted. Oil and gasoline were dropped in 5 5 -gallon drums, protected by sacks 
of rice husks as bumpers, and suspended from multiple parachutes. 

Dropping supplies from the aircraft to the ground raised major problems 
of packaging. Since air-supply techniques were being worked out on a local 
basis in CBI, and no equipment came from the United States before late in 
1943, Indian material had to be used, and proved extremely practicable. A 
container that could withstand the impact of the ground— a considerable force 
even when parachutes were used— was found in the "country basket." This 
was a woven bamboo frame, covered tightly by hessian cloth, with a cradle of 
heavy M-inch rope fastened to the parachute. Experience revealed that the 
baskets should not be too large, so that a number of them could be kicked out 
of the aircraft on one flight over the drop zone. The basket proved so sturdy 



Table 3 — Tonnage Delivered to Northern Combat Area Command by Air: 

April 1943-March 1945 ■ 

Number of short tons 


i car and niontn 







58, 710 

152, 828 

89, 914 

33, 136 

29, 778 



4, 920 

2, 676 












































38, 897 

101, 874 


22, 826 



2, 589 






1, 831 




4, 648 



2, 928 


2, 290 



2, 543 





1, 895 






1, 773 





9, 456 




4, 974 

12, 748 










14, 360 

10, 598 


1, 392 


15, 894 


3, 294 

2, 394 

5, 588 

13, 843 



2, 043 



46, 034 

29, 201 

7, 634 

9, 199 



8, 985 


4, 206 


5, 829 

14, 716 

9, 545 


2, 859 


15, 103 

10, 671 

2, 298 


• Excludes air supply to Galahad. 
Source: NCAC History, App. 6, pp. 16-17. 

and resilient that there was no need to pack rice husks or sawdust round its 
contents. Made by Indian contractors, the country basket was a most useful 
tool, and was used in great quantity. Parachutes, too, were of Indian 



Because parachutes and containers were relatively expensive and scarce, 
free-dropping was used when possible. Rice, salt, animal forage, and various 
ration items for Indian troops lent themselves well to being packed in 35-pound 
burlap sacks which in turn were put into larger sacks. Once labeled with a 
code symbol to indicate the contents, the sack was ready for dropping. 

By ingenuity it was found possible to deliver items that would seem most 
unlikely candidates for airdropping. Chinese units received live poultry and 
pigs as rewards for distinguished action or to mark feast days. Eggs were safely 
dropped. Radios, radio parts, high explosives, field ranges and small carts were 
able, when carefully packed, to survive dropping. 45 

Initial operation of the priorities system tended to overload communications 
facilities. In July 1944, the theater signal officer joined with the G-4 of 
Stilwell's field headquarters to obtain better co-ordination of commercial, SOS, 
and AAF lines. "Top priority for their use after 1500 hours was given to air 
control." Code machines were obtained and used to send messages in the clear. 
The form of messages was also improved. Initially, Signal Corps insisted on 
use of its own style in the arranging of messages, with numbers spelled out and 
the whole text given as one big paragraph with no indentations for numbered 
subheads. As a result, messages on receipt had to be retyped to make them 
usable. Repeated protests finally brought agreement to sending radios in any 
logical readable form. 46 

Problems of Air Supply 

The greater number by far of the troops supplied by air in north Burma 
were Chinese, and Chinese cultural traits in contact with the latest Western 
gadgets produced results which ranged from the sidesplitting to the tragic. 
Americans and Chinese might not always agree on what was humorous and 
what was not, but contacts between the two peoples produced a wealth of 
anecdotes and a number of problems. On one occasion, a Chinese unit was first 
at the drop field when some blood plasma was delivered. The plasma disap- 
peared, and only later was it learned that the Chinese had made their pack 
horses drink it. "But why?" they were asked. "Weren't your horses well?" 
"Perhaps so," the Chinese admitted, "but the medicine made them better!" 

Chinese troops were far easier to supply by air than were U.S. or British 
soldiers. The individual Chinese required but three pounds of rations a day as 
against the American's five pounds. He did not ask the variety in his diet that 
the American demanded. The Chinese was not only extremely careful of his 
possessions, in contrast to the habitually wasteful American soldier, but 
Chinese units had a way of acquiring more, and more, and still more equip- 
ment, while American units could be trailed by what they discarded. Conse- 

45 NCAC History, App. 6, pp. 11-22. 

46 NCAC Air Supply, pp. 21-111 to 31-111. 



quently, "once Chinese Units were fully equipped little was ever heard from 
them except requests for ammunition, rations, medical supplies and certain 
short-lived items of equipment." 

But the Chinese had their failings. Every Chinese unit of any size had its 
liaison unit of an American officer or two and some enlisted men. These 
received their own rations, mail, and supplies by airdrop through the Chinese 
division G-4. Somewhere between the Chinese staff officer and the liaison 
team, American mail, packages, and supplies tended to vanish. In the classic 
Chinese fashion, rations and ammunition were overordered on the grand scale. 
Thus in December 1943, the Chinese 38th Division's requisitions were 280 
percent in excess of its strength. 

Naturally, there were times when even the extraordinary carrying powers of 
the Chinese could not cope with this manna raining down. Supply dumps and 
less formal caches were left behind, and there was great barter with the 
Burmese. When the Yunnan border was approached, pack trains began to 
carry American lend-lease to the Yunnan bazaars, as in the days when the 
Burma Road was hailed as "China's life line." 

Many American service units such as hospitals were attached to Chinese 
units and shared the same dropping fields and airstrips. Sincerely grateful as 
the Chinese were for the medical care they received, many Chinese failed to see 
any connection between it and the food that was dropped for hospital person- 
nel. Consequently, the medicos had to be vigilant, quick, and strong, to claim 
their rations on the instant. 

The Chinese seemed to find it hard to mark dropping fields correctly, and 
having once marked a field were reluctant to mark it again, feeling that every 
pilot in north Burma would know the spot once supplies had been dropped on 
it. Occasionally, Chinese fired on transports coming in for a supply drop, and 
were extremely careless about putting mortar or artillery fire through the air 
space in which transports were flying. For their part the transport crews some- 
times carried small arms and grenades and would use both against areas which 
they believed held Japanese. They were sometimes mistaken in their beliefs. 

Nor did the Chinese ever seem to realize that there was genuine danger in 
the falling packages. The Chinese "stood on the field or rushed out from the 
sidelines to catch parachuted supplies as they floated like manna from the skies. 
Many of these were killed by free dropped rice bags, some crushed and not a 
few actually beheaded." 47 

The basic cause of much of the Chinese soldiers' behavior probably lay in a 
fear that some day this wonderful and incomprehensible flow would end, a fear 
reawakened whenever supplies failed to arrive on time. No Chinese general 
had ever been so mad as to pour out goods and wealth like this! Surely even 
the American barbarians would someday realize that one did not use good iron 

47 (1) NCAC Air Supply, pp. 5 7— III to 63-HI; quotation on pp. 61, 62. (2) Ltr, Dupuv to 
Ward, 12 Sep 52. OCMH. 



to make a nail or a good man to make a soldier. But though the supply ma- 
chine sometimes faltered, it never failed, and the Chinese soldier continued his 
patient, valorous, enduring advance. 48 

Equipment for airdropping was sometimes in very short supply. During 
November and December 1943 it was necessary to fly supplies of rope and hes- 
sian cloth from Calcutta to Assam since fresh stocks were not forthcoming 
through regular channels. Investigation revealed that supplies which had been 
dispatched some weeks before had been sent by river barges. These in turn had 
been stranded during a period of low water. After this experience supplies were 
sent by rail all the way. Parachute stocks on occasion fell so low as to cause 
concern. Such emergencies were met by applying "extreme pressure" on the 
troops to collect, preserve, and return parachutes, for normally no parachutes 
were ever returned from the front. 

Because delivery of the desired items to the proper point depended on ade- 
quate information, quick and accurate communications between the central 
control point for air supply at Ledo and the airfields was necessary. One reason 
lay in the fact that sometimes it was necessary to change the destination of a 
plane while it was in the air. For another, a steady flow of data as to the re- 
quirements of troops was necessary for effective preplanning. Until a teletype 
system was installed in May 1944, a telephone was used, and, because of in- 
herent deficiencies in the Indian system of that date, was unsatisfactory. 49 

The supply effort was not without its price. From January to August 1944 
inclusive, the AAF lost 32 transports destroyed or missing, and had 24 dam- 
aged. 50 The dollar cost was high, for air supply is an expensive operation, 
whose economy is found in its speeding of victory. Even using equipment that 
was of Indian manufacture and so less expensive, SOS estimated the cost of air 
supply as: 

$1,909.65 per ton parachuted 

$ 94.07 per ton free-dropped 

$ 49.61 per ton landed 51 
The cost of parachuting supplies reflects the price of the parachute. To drop 
a ton of supplies using American parachutes called for, among other things, 
fourteen 24-foot parachutes at $72.00 each, a total of $1,012.00. Free-dropping a 
ton of supplies called for $14.75 of drawstring bags and stencil equipment. Ex- 
penses could have been sharply reduced if troops and commanders had co- 
operated in salvaging parachutes, but less than 1 percent was ever reclaimed. 
Parachutes were a great aid in softening some of the miseries of war in the 
jungle, doing service as objects of barter with the Burmese, foxhole lining, 
"tentage, bandages, target panels, towels, handkerchiefs, sheets, and pajamas." 

48 NCAC Air Supply, pp. 57-111 to 63-111. 

49 NCAC History, App. 6, pp. 25-26. 

50 EAC Despatch, App. 10. 

51 NCAC Air Supply, p. 1-VII 



On its own initiative a drop team of the Chinese 50th Division salvaged 998 
British and U.S. parachutes. The saving of $38,000.00 drew a letter of appre- 
ciation from the commanding general to the division commander and his drop 

The cost to the Air Forces of one sortie was about $1,285.00. Total expense 
of delivering 18,300 tons in one month was $12,688,742.25. The major item 
was operating the aircraft, $7,760.00. Next was the expense of parachuting 
2,394 tons, or $4,571,000.00. Landing 10,206 tons cost but $47,049.66, which 
points an obvious moral, since dollars are the bookkeeping symbol for the ex- 
penditure of a given amount of energy, skill, and raw material by the economy 
at a time of great strain. 2 

The B-29 Command Problem 

The decision to place B-29's in China, Burma, and India injected yet an- 
other command problem into an already tangled situation. Chennault and 
Mountbatten both wanted to command the B-29's, confronting Stilwell with 

" NCAC Air Supply, pp. 1-VIIff. 



a delicate problem of intratheater relationships. Brig. Gen. Edgar E. Glenn, 
Chennault's chief of staff, told General Hearn, his opposite number in Stil- 
well's headquarters, that it was a "well known fact that the Chinese have ap- 
plied for and have been refused representation on the Joint Chiefs of Staff [sic], 
the reasons being that the sphere of influence of the JCoS [sic] did not include 
China." Because Glenn understood that the JCS were retaining ultimate con- 
trol of the B-29's, he felt that under the circumstances the command arrange- 
ment would anger the Chinese. He further claimed it was unjust that Chen- 
nault should not be given command of the B-29's and informed Hearn that 
the proposed operations "demand a complete unification of command for 
proper co-ordination." Glenn demanded as well that that the B-29 project be 
completely divorced from SEAC. 53 

Hoping to receive command of the B-29's and approval of his strategic 
views by a presidential order overruling the Joint and Combined Chiefs of 
Staff, Chennault in late January 1944 put his case before the President through 
direct correspondence. 54 He may have believed that events had provided the 
answer to the objections Stratemeyer had raised to his plans in October 1943. 
After Sextant, the Generalissimo had been offered the opportunity of delay- 
ing his portion of the Burma campaign until November 1944. This offer could 
be taken as reason to divert Hump tonnage from the Chinese ground forces to 
the Fourteenth Air Force. In December 1943, Hump tonnage had jumped from 
the 7,300 tons of November to 13,450 tons as the Trident decisions finally 
bore fruit. 

Basically, Chennault's memorandum and letter to the President were the 
familiar Chennault Plan, but with considerable attention to the command 
problem posed by the B-29's. Chennault recommended that the B-29's under 
his command should attack Japanese industrial targets after air supremacy in 
China had been won and Japanese ocean shipping decimated. Fourth priority 
should go to attacks on Japanese Army installations in China proper, Formosa, 
and Hainan. The Fourteenth Air Force's commander stated that if the Chinese 
Army attacked the Japanese positions on the Yangtze River after the Four- 
teenth Air Force had vigorously bombed them and injured the Japanese line of 
communications the Japanese would soon begin to withdraw. Consistent with 
his earlier proposals, though he stressed the need for improving the line of 
communications to his airfields and estimated the needs of that project in terms 
of Hump tonnage, Chennault would provide no Hump tonnage for the Chi- 
nese Army. Accordingly he wrote: 

8. The objection, that the Japanese ground forces can capture and destroy the East China 
air bases, has always been made to previous proposals for a China air offensive of this type. 
This objection is founded, however, on experience in a period when the Chinese armies were 
entirely without air support. It also ignores the most striking fact of the war in China, that 

53 Rad M 27 JB, Glenn to Hearn, 8 Jan 44. Item 1606, Bk 5, JWS Personal File. 

54 Ltr, Chennault to President, 26 Jan 44. Hopkins Papers. 



Japanese forces have never succeeded in penetrating Chinese territory to a distance of more 
than about 100 miles from a major supply line. This was so even when the Japanese enjoyed 
air supremacy in China. The situation has now been radically changed by the transfer of air 
superiority to the Chinese side. In the first place, as has been noted already, traffic on the 
major Japanese supply line within China, the Yangtze River, has actually been interdicted 
experimentally for a short period of time. Almost all Japanese supply lines within China 
are water borne, and can no longer be considered dependable. In the second place, the recent 
battle of Changteh, the first in which Chinese troops have had well organized air support, 
showed that with this assistance the Chinese armies are quite capable of resisting and throw- 
ing back a serious Japanese advance. 

9. It is not denied that the Japanese might perhaps capture certain of the East China air 
bases if they organized a major offensive, employing large numbers of first class troops and 
an overwhelming strength in the air. But organizing such an offensive would be self-defeat- 
ing. By concentrating their effort in China, the Japanese would inevitably risk rapid loss of 
their new empire in Southeast Asia and the Southwest Pacific. 55 

Chennault also pressed his claims with Arnold. He told Arnold that the 
cost of B-29 operations from the Cheng-tu area would be prohibitive because of 
radar-directed Japanese fighters. Again, Chennault insisted that "the commander 
of the air forces in China must be given complete command and control of 
all such air forces and supporting services as are based in or are operating from 
China." 56 Whatever Arnold's views on command may have been, he was 
wholeheartedly behind Matterhorn, the plan for basing B-29's in India, 
staging them through China, and using them to cripple Japan's steel industry. 
This was apparent in his comments on a letter from Wedemeyer which ques- 
tioned the possibility of a major bombing offensive from China. Wedemeyer 
urged that all Hump tonnage go to the Fourteenth Air Force. On receiving 
Wedemeyer's views, Arnold promptly passed them on to Marshall with the re- 
mark that they were just another attempt to discourage the long-range bomb- 
ing of Japan, and reminded Marshall that not long before efforts had been made 
to discourage "our" flying 10,000 tons a month over the Hump. 37 

It is not known whether the President replied directly to Chennault's Jan- 
uary 1944 proposals. However, one may surmise that many things had hap- 
pened since May 1943 to change Chennault's position vis-a-vis the President. 
At the time of the Trident Conference, May 1943, the President had been 
paying much attention to China. China was an issue in American domestic 
politics, as speeches in the Senate made clear. Now, in 1944, American atten- 
tion was increasingly drawn towards Europe. Everyone in high position knew 
that the cross-Channel assault was scheduled for spring and that the whole war 
would revolve on it. Against this tremendous fact, even the colorful and dy- 
namic Chennault could not compete. Since the President's personal attention 

55 Plan of Air Operations in China, 1 Jul-Dec 44. Folder, Chennault Air Plan, CT 39, Dr 1, 

56 Ltr, Chennault to Arnold, 26 Jan 44, sub: Matterhorn Project. Bk IX, Hopkins Papers. 

" Memo, Lt Col George C. Cox for Col Frank McCarthy, 28 Jan 44; Memo, Arnold for Mar- 
shall, 29 Jan 44, sub: Bombing of Japan. Case 282, OPD 381 China, A47-30. 



Chart 5 — Tonnage Shipped From India to China by Air : 1944 



was being given to pending operations in Europe, he was no longer desiring 
to set Hump priorities. 

What attention Roosevelt did spare to air matters in China went to the 
B-29 project. By an irony of history, Chennault and his friends now occupied 
the spot Stilwell and Marshall had filled in May 1943— that of arguing against 
a project close to the President's heart. 

Therefore, the whole of Chennault's October-January proposals for 1944 
was not approved. It remained politic to give the Fourteenth Air Force ap- 
proximately half of the Hump tonnage being received in China. ( Chart 5 ) The 
careful studies of the line of communications to the east China bases, made by 
the Fourteenth Air Force and the SOS, bore fruit in a series of theater projects 
to improve the eastern line of communications. 58 Theater headquarters ap- 
proved Chennault's proposals for the organization of the reinforcements ar- 
riving for him. In late December 1943, the 68th Composite Wing under Col. 
Clinton D. Vincent was set up for operations in east China; the 69th Com- 
posite Wing under Col. John Kennedy, for operations in the west. In January, 
eighteen P-40's and thirty-three B-25's arrived to reinforce, but bad weather 
limited operations. 59 

Admiral Mountbatten was also concerned over the place of the B-29's in 
the Allied command structure in Asia. Since his arrival in Asia, Mountbatten 

58 Joseph Bykofsky, The History of Transportation Service in China, Burma, and India, in 
World War II (Bykofsky MS). OCMH. 

59 Organizational structure of the Fourteenth Air Force is explained in Fourteenth AF History. 



had been confronted with delicate and involved command problems. There 
was the problem of Mountbatten's relations with his three British command- 
ers in chief, one of whom had been almost openly hostile; there had been the 
problem of Mountbatten's relation to Stilwell, who was extremely sensitive 
and highly suspicious of Mountbatten; there had been the problem of inte- 
grating the air forces into EAC. Mountbatten's remedy for this had been to 
blend tact with firmness. Where it seemed advisable he asserted his authority; 
on other fronts, he waited to see how the cards were falling. 60 

Now came the proposal to place still another command within SEAC, one 
whose logistic needs would have to be met through the port of Calcutta on 
which Mountbatten depended. Support of the B-29's could well mean a diver- 
sion of his resources to achieve ends that would contribute nothing toward 
accomplishing his directives. 61 Consequently, he objected to independent B-29 
operations in his theater and offered suggestions to the CCS, which if accepted 
would have placed the B-29's at his disposal. Agreeing that the B-29's should 
operate under the general direction of the JCS, he suggested that missions be 
assigned by the chiefs of staff organization responsible for the theater in which 
the B-29's were based (for SEAC, the British Chiefs of Staff were responsible). 
Orders from these would go to the commander of the B-29's and to the thea- 
ter commanders in whose areas the missions would be executed. The B-29 
commander would be required to co-ordinate each mission with each theater 
commander concerned. In effect, under such an arrangement, each B-29 mis- 
sion would require the JCS's obtaining the previous assent of the British Chiefs 
of Staff, Mountbatten, and the Generalissimo. Mountbatten saw the justifica- 
tion for these suggestions in the probable clash in priorities between Matter- 
horn and other projects in his theater. 62 

The initial CBI Theater orders setting the place of the B-29's in the theater 
command structure directed that XX Bomber Command be placed under the 
command and control of Stilwell, who in turn was designating Stratemeyer to 
exercise it on his behalf. 63 Stilwell did not want Chennault to command the 
B-29's for fear he would base them at Kweilin and the other east China air- 
fields to raid Japanese shipping. Stilwell believed such a move would provoke 
an immediate Japanese reaction and feared the Chinese could not defend the 
airfields. A successful defense, he now estimated, would call for fifty rebuilt 
Chinese divisions, which were not at hand. 64 

6n ( 1) Extracts, SAC's Personal Diary, 12, 22 Dec 43, 1 Jan, 16 Feb 44. SEAC War Diary. 
(2) Ltr, Mountbatten to Field Marshal Sir John Dill, 26 Jun 44; Ltr, Wedemeyer to Marshall, 9 
Jul 44. Item 70, Folder 57, OPD Exec 10. 

61 Rad RE-338, Egan to Stilwell, 24 Feb 44. Item 48, Bk 6A, JWS Personal File. 

62 Rad DIAL 29, SEACOS, 25 Feb 44. SEAC War Diary. 
61 GO 13, Hq USAF CBI, 30 Jan 44. 

64 (1) CM-IN 8578, Stratemeyer to Arnold, 14 Jan 44. (2) Rad Aquila 31, Stilwell to Mar- 
shall, 23 Jan 44. Item 1670, Bk 5, JWS Personal File. (3) Memo, Timberman for ACofS OPD, 
2 Nov 43, sub: Bomber Offensive from China. Case 192, OPD 381 CTO, A47-30. 



Eleven days after Stilwell put the B-29's under Stratemeyer, Brig. Gen, Ken- 
neth B. Wolfe, commanding the XX Bomber Command, called on Stilwcll to 
discuss the command problem. The two men agreed chat command of the 
B-29's should be kept from both Chennault and Mountbatten. Stilwell be- 
lieved that Stratemeyer, because of the integrated nature of EAC, was vulner- 
able to pressure from Mountbatten, and even more, feared that Stratemeyer con- 
remplated arrangements which would give Chennault practical control of the 
XX Bomber Command. Therefore, with General Wolfe's concurrence and in 
accord with a suggestion from Marshall, Stilwell set up a new command struc- 
ture on February 15, 1944. 65 

The order Sri 1 well issued stated that "following general directives from the 
United States Joint Chiefs of Staff * the XX Bomber Command would be under 
Stilwell's "'direct command and control. " Logistical support, so far as it lay 
within the capabilities of the local air force commanders, was split between 
Stratemeyer and Chennault on a geographic basis, the former in India, the lat- 
ter in China. Where B-29 operations in Southeast Asia and China were con- 
cerned, Stratemeyer was to consult with Wolfe, then offer his recommendations 
to Stilwell. Chennault was to do the same for B-29 operations irom bases in 
China. In effect, this meant that Wolfe would direct the XX Bomber Com- 
mand in operations along the lines laid do*n by general 

The manner in which these orders had been framed is most indicative of 
the personal relations that obtained among several of the China-Burma-India 
Theater commanders on the eve of the theater's most inrense activity, and so 
was significant. But on 28 March theJCS brushed aside the contentions of the 
local commanders and decided that the China-based B-29 ? s would be com- 
manded from AAF headquarters in Washington, In effect, they would be com- 
manded like a fleet at sea which might base now at this island, and now at that, 
bur would not be tied to any one theater/ 10 Stilwell and Chennault, therefore, 
hud their authority circumscribed in that this gftatt new engine of war would 
draw on their reserves, would affect events in their theater, for which rhey 
would later be held responsible, but would not be at their disposal. 

Logistical Problems of the B-29 ! s 

In retrospect, Maj. Gen. Vernon Evans, who had a wide and varied experi- 
ence in CBI as Deputy Chief of Staff Rear Echelon (the New Delhi headquar- 
ters), and later Chief of Staff and Theater Commander, Tndia-Burma Theater, 
remarked that the support of XX Bomber Command was among the major 
logistical problems of CBI Theater. " Since CBI Theater had built roads and 

<* (1) Stilwell Diary, 11 Feb 44. (2) Note, 

W Wesley Frank Craven and James Lea Care, 
Pacific: MATTERHQRM to Nagaiaii t Jm« 
cago Press, 1953), pf- 45-52 

« T Lnrerv with G* 



pipelines across the jungle and the mountains, had equipped and supplied 
Chinese troops, and nurtured two air forces, it is plain that General Evans was 
comparing preparations for and support of the XX Bomber Command with 
logistical problems of the first magnitude. 

CBI Theater had to build five airfields in India for the B-29's, complete 
with facilities, and link them to the oil tanker terminals by a pipeline; arrange 
for the building of airfields in China; supply common-user and Air Forces 
items, both in India and in China; furnish higher echelon maintenance; fly 
gasoline and supplies over the Hump to Cheng-tu. The last was a tender point. 
Though Matterhorn was supposed to be fully self-supporting, from Febru- 
ary to October 1944 ATC had to fly 17,931 tons over the Hump to Cheng-tu. 
This tonnage was a very heavy drain on the meager resources available for U.S. 
operations in China, and equal to the tonnage flown into China for the Chinese 
Army from May 1942 to October 1944. 68 

CBI began its share in B-29 airfield construction in India by supplying 
trucks and heavy equipment from other projects to supplement the Indian 
forces extending the existing runways. The U.S. engineer battalions allotted to 
the task were slow in arriving and it was necessary to divert two aviation en- 
gineer battalions from the Ledo Road. By 18 March, SOS, CBI, could report 
that the fields at Kharagpur and Chakulia were "barely operational" and on 
schedule. On 30 June four fields were ready, with the fifth, Kalaikundah, de- 
layed because an engineer aviation battalion was diverted for forward airstrip 
building. 69 Because oil supply by tank car would have sorely strained transport, 
it was decided to lay pipelines from the oil terminal at Budge-Budge to the 
fields. \r&e Map T. Jj By 23 February 1944, Indians and Americans working to- 
gether strung 100 miles of pipe, and by 13 March pumping to the storage tanks 
at Kharagpur and Chakulia could begin. 70 

In China, after initial difficulties over finance were overcome, the Chinese 
Government went forward with a tremendous airfield construction program. 
About 200,000 peasants were mobilized, some of them volunteers, most con- 
scripted. To these were later added 100,000 more, plus 75,000 contract laborers. 
American and Chinese engineers laid out the runway extensions, for as in India 
the Cheng-tu B-29 fields were existing fields, extended and improved. Using 
hand implements and pulling rollers by sheer muscle power, the peasants fin- 
ished four B-29 fields and three fighter strips by 1 May 1944. The Chinese War 
Area Service Corps furnished housing and food, for XX Bomber Command 
tried to keep the number of Americans at Cheng-tu to a minimum to lessen 
the burden on the Chinese economy. 71 

68 History of CBI, Sec II, Ch. Vie. (2) Briefing Data Prepared for Maj. Gen. Patrick J. Hurley. 
Hurley Papers. (See Bibliographical Note). 

69 Rpts, Gen Covell, CG, SOS USAF CBI, to Somervell, 4, 21 Jan, 2 3 Feb, 18 Mar, 17 Apr, 
21 Jun, 19 Jul 44. OCMH. 

70 SOS in CBI, pp. 459-60. 

71 History of CBI, Sec. II, Ch. Vie. 




When Stilwell entered the field in December 1943, an important prereq- 
uisite of victory was at hand— control of the air. With care and foresight, he 
had armed and trained two Chinese divisions, was preparing a third, and had 
secured solemn guarantees of untrammeled command. The rapid growth of 
U.S. air power, the increasing knowledge of what air power could do, gave 
Stilwell by December 1943 an advantage that his adversary, General Tanaka, 
could not hope to match. Hundreds of miles ahead of Stilwell the bombers 
ranged, harassing Tanaka's supply lines. Unrestrained, the reconnaissance 
planes peered up and down the valleys. The fighters came on call to blast ma- 
chine gun nests and entrenchments. Endlessly, the transports sailed overhead, 
cascading rice, bullets, and bandages to Stilwell's Chinese. Liaison planes now 
evacuated the wounded from the jungle, where once their injuries would have 
meant slow death. 

In only one way did the new strength and potentialities of U.S. air power 
positively handicap Stilwell and CBI Theater in the campaigns of 1943-44. 
The XX Bomber Command's B-29's, with their demands on Hump tonnage 
and theater facilities, conflicted directly with Stilwell's still unrepealed though 
practically abandoned mission of reforming the Chinese Army. That the B-29's 
were flying from China would be to the Japanese one more reason for a 
major effort there, yet B-29 demands on Hump tonnage made it even more 
unlikely that anything effective would be done to stop the Japanese once they 

With the conclusion of this period of planning and preparation, which may 
be taken as roughly October-December 1943, the Allied war in China, Burma, 
and India proceeded by force of geographic circumstance in two very distinct 
compartments. On the India-Burma side of the Hump, Stilwell was fighting 
his own campaign, to break the blockade of China. For many months, events 
in Burma and India proceeded without affecting those in China, and Stilwell 
left the conduct of the war in China Theater to its Supreme Commander, Gen- 
eralissimo Chiang Kai-shek. As the campaign in Burma moved to its climax, 
the center of interest shifted to China, where events began to crowd ever faster 
on one another. But the two campaigns moved separately, on either side of a 
gigantic mountain range. Though they affected one another, they were related 
only at the highest level; the daily ebb and flow of battle might have been in 
two separate wars. The campaign in Burma was the first to reach the critical 




Breaking the Stalemate in 
North Burma 

A great semicircle of Allied forces stood around Burma in December 1943. 
With the exception of Stilwell's forces, which had breached the mountain 
barriers in north Burma, Allied forces were in positions that roughly conformed 
to the natural defenses that ring Burma in a long Gothic arch with its apex 
toward the north. The forces on the west or Indian side were under SEAC and 
Mountbatten; those to the east or Chinese side, under the Chinese Expedi- 
tionary Force, Gen. Wei Li-huang commanding. In the southwest, in the 
Arakan area along the Bay of Bengal, was the British 15 Corps (5th and 7th 
Indian Divisions and 81st West African Division). Since November 1943 they 
had been driving in the forward positions of the Japanese 55th Division in order 
to seize an area from which, with amphibious su pport, they might take the 

airfields and the port of Akyab on the Arakan coast. (Map 5*) 

About 300 miles to the north, SEAC had its 4 Corps (17th, 20th, and 23d 
Indian Divisions). Here there was only patrol activity, but in early 1944 the 
corps was scheduled to take the offensive into Burma to keep Japanese forces 
from moving north to stop Stilwell. 4 Corps faced the Japanese 33d and 31st 

In the Hukawng Valley, Stilwell had the Chinese 38th Division, with 
elements of the 22d Division coming forward. They were fighting the Japanese 
55th and 56th Regiments, 18th Division, and were driving across north Burma to 
break the blockade of China. 

On the China side, on the long Salween River front, were the eleven 
divisions of the Chinese Expeditionary Force. Opposite them the Japanese had 

*For explanation of enemy unit symbols, see note |82(3). page 42] above. 



their 56 th Division. The Japanese deployment left one Japanese division in 
reserve, the 54th. 1 

The terrain feature on which the campaign, seen in its largest sense, 
currently focused was the town of Myitkyina. Lying in the center of north 
Burma, at the southern tip of the mountain range or Hump over which the 
transports flew to China, and possessed of road and rail links with the rest of 
Burma and with China, Myitkyina had strategic advantages to which the Joint 
and Combined Chiefs attached great importance. Were Myitkyina and its air- 
fields in Allied hands, the transports of the ATC could fly a lower, broader 
route to China. Were Myitkyina in Allied hands, the Ledo Road and its com- 
panion pipelines could link with the prewar communications net of North 
Burma, Myitkyina would become a great supply center, and the end of China's 
blockade would be at hand. 

In their current position around the obscure wrecked villages of Yupbang 
Ga, Sharaw Ga, and Ningbyen, Stilwell's Chinese were on the Burman side of 
the mountains that separate Burma and India. They were at the north end of 
the long corridor of the Hukawng and Mogaung valleys whose southern exit 
is within easy march of Myitkyina, and the Irrawaddy valley that forms so 
much of central Burma. The principal barrier between Stilwell and Myitkyina 
was the three regiments of the Japanese 18th Division. These skilled veterans 
under the competent leadership of General Tanaka could be counted on to 
make good use of the several dominant terrain features that lay between 
Stilwell's troops and the streets, houses, bazaars, and temples of Myitkyina. 

The Sextant decisions of the Combined Chiefs of Staff in December 1943 
required that Stilwell occupy upper Burma in spring 1944. To carry out this 
order Stilwell would have to break the stalemate that had developed in the 
northern end of the Hukawng Valley. These were the elements of the stalemate: 
The three battalions of the 112th Regiment, Chinese 38th Division, had been 
sent forward into the north end of the Hukawng Valley to hold the line of the 

1 (1) The strength of the Japanese forces during the winter of 1943-44 caused discussion 
among Allied leaders at the time, a discussion reflected in some postwar accounts. The Japanese 
took Burma with four divisions, the 18th, 33d, 55th, and 56th. In summer 1943 they added a fifth, 
the 3 1st Division. In December 1943, while trying to persuade the Chinese to attack across the 
Salween into Burma, Stilwell used the estimate of five Japanese divisions. He erred, in that the 
54th Division had entered Burma in fall 1943, but had not yet been identified. The Generalissimo 
in December told Stilwell that there were eight Japanese divisions present. As noted in Chapter I, 
above, reports were circulating in October 1943 that the original Japanese force of four divisions 
had been reinforced by four more; the Generalissimo's estimate may have reflected these reports. 
During the months of January and February 1944, the 2d Division entered Burma by rail, while the 
15th Division arrived during the three-month period of January-March, moving by foot and truck 
over the Kengtung-Takaw road. This made a total of eight Japanese divisions in Burma by March 
1944. The statement in the Mountbatten Report, Part B, paragraph 66, that "by January"— pre- 
sumably by 1 January 1944— the Japanese forces in Burma had been increased to eight divisions 
cannot be reconciled with the information in the SEATIC bulletins and Japanese studies. If one 
allows for the period of many weeks needed for a Japanese division to close at its Burmese station, 
the Mountbatten Report necessarily implies that leading elements of the 2d and 15th Divisions 
entered Burma in October and November 1943. Statements to that effect do not appear in the Jap- 
anese sources used by the authors, Japanese Studies 89, 133, and 134. (2) SEATIC Bull 240. (3) 
Mountbatten Report, Pt. B, pars. 32-53. 



Tarung Hka and Tanai Hka. The Tarung, flowing from north to south, enters 
the westward-flowing Tanai about six miles southwest of the village of Yupbang 
Ga. The rivers lay in the path of the engineers building the Ledo Road; it was 
essential to hold their crossings. Moving forward on this mission, the battalions 
of the 112th had met the outposts of the Japanese, elements of the 55th and 
56th Regiments, 18th Division. 

The Chinese battalions were soon cut off as the Japanese used their cus- 
tomary device of encirclement. The 1st Battalion at Yupbang Ga, the 2d 
Battalion in the Sharaw Ga area (both on the Tarung), and the 3d Battalion at 
Ngajatzup, about twenty-five miles southwest of Yupbang Ga in the Taro 
Plain, were all surrounded, relying on air supply, and apparently powerless to 
cut their way out. Nor had the initial attempts to relieve them been successful. 
The commander of the 38th Division, General Sun Li-jen, had tried unsuccess- 
fully in mid- December to relieve the 1st Battalion at Yupbang Ga. He used 
reinforcements for whose presence he was in considerable measure indebted to 
a visit by the Generalissimo and Madame Chiang two weeks earlier. 2 

The Chiangs Visit Their Troops 

On 30 November 1943 the headquarters of the Chinese Army in India was 
visited by Generalissimo and Madame Chiang on their way back from Cairo; 
they still believed the Allies were in accord on Burma operations. While the 
Generalissimo spoke with Generals Sun and Liao, Madame Chiang asked 
General Boatner if the Chinese troops had not failed to bypass the Japanese 
and go ahead. Boatner shared her impressions. Asked by Madame Chiang to 
explain, Sun replied that he had no reserves available at the front. On Boatner's 
asking if the arrival of these reserves would set him to enveloping the Japanese, 
Sun concurred. Madame Chiang again raised the issue, and again Sun said yes. 

Then the Generalissimo called Boatner before him to receive his views, 
Madame Chiang interpreting. The Generalissimo asked that his expressions be 
passed on to Stilwell. Boatner reported the Generalissimo as saying: 

. . . our forces were at a big disadvantage, that supply was most difficult, that no road was 
available for rapid troop movements and the Japanese had every advantage to include large 
forces in our immediate front. He stated that we should not provoke a large-scale battle with 
the Japanese and we should not cross the Tarung-Tanai River until February, because at that 
time the British would move to the South and the Chinese would move from Yunnan. He 
stated that the Chinese force used to hold the present river line should not exceed one 
regiment. He explained that this force was desirable because we had only six regiments and 
if two were cut off by the enemy we would have only four regiments left. 3 

General Boatner replied that at most two Japanese battalions faced the 38th 

2 Overlay, Hq 5303d (Prov) Combat Trs, 13 Nov 43; Overlay, 4 Dec 43. NCAC Files, KCRC. 

3 Memo, Boatner for Stilwell, 2 Dec 43, sub: Rpt on Conf with Generalissimo and Mme. 
Chiang, 30 Nov 43. Stilwell Documents, Hoover Library. 



Division, that air supply kept the Chinese well fed and well armed, that the 
movement forward was, properly speaking, not an offensive at all but made 
necessary by the progress of the Ledo Road. Boatner also explained the system 
of liaison and command— that he, Boatner, simply passed on StilwelPs orders. 
Then the gathering paused for dinner. 

The minutes of the conference suggest that in the light of the First Burma 
Campaign's experiences Boatner was alarmed by the Generalissimo's statements 
before Sun and Liao that the defensive should be adhered to and only one 
regiment risked. As soon as the Generalissimo's party sat down to dinner, the 
Generalissimo again stated that in his opinion only one regiment should 
defend the river line. Asked what would happen if that regiment were not 
enough, the Generalissimo replied: "Well, of course you will have to think up 
a method (Hsiang I Ko Pan Fa)." Boatner acted at once, and: 

. . . immediately told Madame Chiang that I considered the Generalissimo's statement in 
reference to using only one regiment of the utmost importance; that if he told his division 
commanders that only one regiment would be used and if thereupon he, the Generalissimo, 
left the area, the situation might change abruptly and a catastrophe result before he could 
make necessary changes in orders. I quickly added, again to Madame Chiang, that this was 
of utmost importance and the Generalissimo must realize that many of his officers took his 
views and wishes as explicit orders, and that such would severely handicap future operations 
in this sector. Upon receiving Madame Chiang's interpretation, the Generalissimo quickly 
replied that these were not orders, only his views, and that you [Stilwell] had complete 
authority. I then asked Madame Chiang if she would make certain that the Generalissimo 
would make this crystal clear to Generals Sun and Liao. The Generalissimo agreed. 

After dinner the Generalissimo told Sun and Liao in a very formal manner 
that his remarks on operations were simply his personal views and that Stil- 
well's orders were to be carried out in full. The meeting had been friendly and 
cordial throughout, thought General Boatner, and he so reported. 4 

On learning of this conference, Stilwell told Madame Chiang that the stale- 
mate had been caused by General Sun's failure to maneuver aggressively. (It 
will be recalled that the tendency was to underestimate the number of Japanese 
then present around Yupbang Ga.) The discussion, partly personal and partly 
by radio, ended with Stilwell's assurance that reinforcements were coming, and 
Sun's solemn promise to Madame Chiang that he would indeed bypass the 
entrenched Japanese when his reinforcements arrived. Stilwell capped the 
episode by writing a letter to Sun asking him to make his promise to Madame 
Chiang a matter of record. 5 

Stilwell then told Boatner that Sun was free to use the whole 38th Division 
and could move up a regiment of the 2 2d Division to a reserve position if he 

4 Memo cite J n. 3.1 

3 Ltr, Stilwell to Sun, 11 Dec 43, sub: Unsatisfactory Methods Employed During Present Opns; 
Ltr, Stilwell to Sun, 9 Dec 43, sub: Accomplishment of Promise Made to Mme. Chiang 30 Nov 
43. Folder, Combat Rpt Fwd Ech and Ln O's, Albacore Hist Files, NCAC Files, KCRC. 



wished; the latter request Boatner had earlier refused because he wanted to 
train the 22d Division for a few months more. 6 

Almost immediately, General Sun tried to relieve the beleaguered 1/1 12th 
at Yupbang Ga. On 15 December the l/ll4th, supported by the 6th Battery, 
38th Division Artillery, attacked. Firing the first preparation of the campaign, 
the battery was thought to have driven away the crews of three of the four 
Japanese machine gun positions separating the two battalions. The Chinese 
waited ten minutes, then gingerly probed the Japanese positions with a small 
patrol. Fired on by the remaining machine gun, the Chinese withdrew to their 
own lines. The Japanese then reoccupied and strengthened their positions so 
that later attacks found them much more formidable. 7 

Yupbang Ga 

Taking up his duties as Commanding General, Chinese Army in India, Stil- 
well arrived at Shingbwiyang, in the Hukawng Valley, on 21 December. That 
day and the next he spent with General Sun and the several staffs in examining 
the local situation. In the Yupbang Ga-Sharaw Ga-Ningbyen area, the ser- 
pentine Tanai Hka, flowing roughly from southeast to northwest, most shapes 
the local topography. Almost immediately south and southwest of the Tanai 
is the hill mass of the Wantuk Bum. Flowing almost due south until it meets 
the Tanai is the Tarung. The two rivers thus form the eastern and southern 
boundaries of a small terrain compartment. The traveler or the army that wants 
to leave the compartment on the south must use the Kantau ford; the eastern 
exits are the fords at Yupbang Ga and Ningbyen. To get his campaign under 
way, Stilwell would have to pry the Japanese grip from the Chinese units at 
the Tarung Hka crossings at Yupbang Ga and Ningbyen. Fortunately for him, 
the Japanese had omitted to g uard the Kan tau ford, so that the exit to the 
south across the Tanai was open. \(See Map 4.)\ 

The Japanese positions seemed to Stilwell and his staff to be along the 
Tarung Hka with their left flank resting somewhere near the hills of the Wan- 
tuk Bum. Thus, their line ran north and south, roughly parallel to the Tarung, 
and the unguarded Kantau ford seemed to open an obvious opportunity to at- 
tempt an envelopment of their left or southern flank. If the Chinese could 
move a force across the Tanai at Kantau, then send it moving eastward along 
the Tanai's southern bank, they might well succeed in placing themselves in 
the Japanese rear and cutting off Tanaka's force north of the Tanai. Such a 
stroke might well result in disaster for Tanaka's 55th and 56th Regiments and 
decide the campaign in its first months. 

6 Sun had complained to Stilwell that the orders he received even specified the units he was to 
use. Rad, Stilwell to Boatner, 14 Dec 43; Rad RELOT G 335, Boatner to Ferris and Hearn, 6 Dec 
43. Items 1518, 1504, Bk 5, JWS Personal File. 

7 Comments by Col Dupuy on NCAC History, 1, 29 (hereafter Dupuy Comments); Notes by 
Col Dupuy on draft MS (hereafter, Dupuy Notes); Ltr, Dupuy to Ward, 24 Sep 51. OCMH. 



Sun's initial proposals were to send the 3/ll4th, 38th Division, south 
across the Tanai at Kantau ford to cut around the Japanese left, while two ad- 
ditional companies tried to swing around the northern end of the Japanese line 
by crossing the Tarung north of Sharaw Ga. Stilwell objected that the pro- 
posed forces were too small. In lieu of the attempted envelopment of the Jap- 
anese northern flank by two companies, he substituted a strong attack toward 
Yupbang Ga, which, after breaking the Japanese center and relieving the 
l/112th, would hook north and envelop the northern fragment of the Jap- 
anese line. The force sent to move around the Japanese left was finally set at 
a regiment. 8 

The commander of the Chinese 114th Regiment was put in command of 
the operations around Yupbang Ga, since his unit had been chosen to make 
the effort, with the 112th Regiment co-operating as best it could. Supported 
by the 5th and 6th Batteries, the l/ll4th deployed with companies in line. 
Four Japanese strongpoints in a rough square barred them from the 1/1 12th. 9 

Preceding the attack, which was launched on 24 December, another artil- 
lery preparation was fired, with great accuracy, for only thirty yards separated 
the two forces at some points. When the Chinese went in at 1000, the Jap- 
anese held stoutly, because this time there was no element of surprise. The 
Chinese had waited five minutes after the fire lifted before attacking and had 
blown a bugle in the accepted Chinese practice. The Japanese fought with their 
accustomed stubbornness and counterattacked thirty minutes after the Chinese 
companies in the attack at last made contact with each other at 1300 in the Jap- 
anese position. Small parties of Japanese held out in foxholes and dugouts. 

At 1500 the Chinese battalion commander, Maj. Peng Ke-li, enveloped the 
Japanese right flank, and at 1515 another element made contact with the be- 
sieged battalion, which had remained in its lines during the fight. By dark the 
smallest Japanese pocket had been surrounded, and it was wiped out the next 
morning though one dogged machine gunner was still firing at 1000. The sur- 
vivors of the other pocket fought their way clear during the night. The enemy's 
defense had been active. Every night patrols were out raiding, and the Jap- 
anese still held the west bank of the stream. 

This little battle, bitterly fought at close quarters, made a great impression 
on the men of the 38th Division. Many of them were new, had heard stories of 
the 38th Division's successes in the First Burma Campaign, and now felt that 
they too could beat the Japanese. The Chinese soldiers talked of it over and 
over again during the rest of the drive on Myitkyina; the first victory is never 
forgotten. 10 

8 Note in Stilwell's hand, written between 8 and 11 January 1944. JWS Misc Papers. 

9 (1) Dupuy Comments. (2) The Stilwell Papers, pp. 273-74. (3) Stilwell Sketch, NCAC 
History, I, 30B. (4) Dupuy Notes. 

10 (1) Photostat 203, Stilwell Documents, Hoover Library. (2) Stilwell Sketch, NCAC His- 
tory, I, 30B. (3) Dupuy Comments. (4) Major Peng received the Silver Star for this action. GO 
46, Hq USAF CBI, 15 May 44. 

-3? A- Pf~ 



The Japanese, too, were impressed by Yupbang Ga: . . the unexpected 
stubbornness of the Chinese troops in the fighting around Yupbang Ga led the 
Japanese to believe that troops that faced them wetc far superior in both the 
quality of their fighting and in their equipment to the Chinese troops they had 
been fighting in China for years, Too, after witnessing a spectacular [supply] 
airdrop of Allied forces the Japanese realized that the fighting power of the 
American -Chinese forces was not to be underestimated," 11 

The success pleased Stilwell, but a warning of a major Japanese threat may 
have spoiled his Chrisrmas Day. Accepting at face value intelligence that he 
was to be attacked by the Japanese 1 8th and 3 1st Divisions, he called General 
Sun into conference and told the Chinese commander of his plans to meet so 
grave a threat. Stilwell planned to seize Taihpa Ga with the 38th Division, 
then fortify it T while the 2 2d Division did the same in the Taro Plain. Chih 
Hui Pu was to speed the arrival of the 65th Regiment, 22d Division. Thus, for 
the next few days Stilwell's intent was to seize good defense positions, "then 
let Japs attack." u 

The initial success at Christmas did not take Yupbang Ga from the agenda, 
The Japanese between the 1/1 12th and the river still held > blocking the cross- 
ing. Three battalions were to be used in driving them out, the l/ll4th, 
2/ll4th, and 1/1 12th, supported by the 5th and 6th Batteries. Facing them 
were at least the 4th and 6th Companies, 2/ 55th. with machine guns attached, 
holding three strongpoints some 300 yards northwest of the river, and three 
more along its bank. 

Attacking from the north, the three Chinese battalions began the operation 
at 1000, 28 December. By 1430 the 1 /112th on the right flank had swept in 
behind the three outer strongpoints to take one of those along the river. The 
position forming the left flank of the Japanese outer defenses fell at 1400; then 
the Chinese found that this point and its immediate neighbor were linked by 
tunnels through which the Japanese freely moved. Little progress was made 
until the Chinese who had assaulted these defenses bypassed the Japanese sup* 
potting position and went on to the river. This stroke seemed to force the 
Japanese from the three outer positions. In the meantime the 1/1 l4rh broke 
the resistance at one of the points on the river, which had been somewhat iso- 
lated from the other Japanese positions. A counterattack that night from one 
of the two remaining strongpoints failed, and the point itself was wiped out 
the next morning. 

The survivors of the Japanese companies split into small groups which 
held out for several days. There was a fluid situation, with "firing all around." 
The last Japanese strongpoints in a sort of anticlimax, held out till 13 January . 

" Tana k a, Japanese Comments, Sec. II. Draft chapters on the north Burma fi^KmHwe^eserit 
tp General Headquarters, Far East Command, to obtain Japanese comments. <See lBibhograPhicaFl 
Note.) The comments were returned in a manuscript, hereafter cited as Japanese Comments, which 
is in OCMH. 

11 Stilwell Diary, 25 Dec 4! 



In the first attempt to clear it on 5 January, the Chinese met land mines, which 
inflicted forty casualties. On the next day an attempt with three companies fol- 
lowing an artillery preparation failed when one company commander attacked 
ahead of time. The position was penetrated but the Chinese suffered such 
losses that they had to fall back. By good fortune, the last attack was made as 
the Japanese were finally pulling back across the river and yielded an unex- 
pectedly large dividend of casualties. 

The Chinese units trapped to the north, in the Sharaw Ga area, were re- 
lieved by elements of the 113th Regiment on 31 December 1943, for with the 
Chinese successes around Yupbang Ga the Japanese position to the north was 
plainly untenable. The line of the Tarung Hka was now firmly in Chinese 

At the south end of the line, in the area where the Tarung flows into the 
Tanai, and where Sun planned to envelop the Japanese left, the envelopment 
had not progressed with the speed Stilwell desired. The 3d Battalion of the 
114th, later followed by the 2d Battalion, marched south to the Kantau ford 
and crossed the Tanai in the second week of December, moving in column of 
companies. Its 8th Company stayed at the Kantau ford; the 7th Company 
halted at the second bend of the river from Kantau and dug in. To the very 
skeptical Stilwell, Sun claimed that he had ordered the 6th Company to pro- 
ceed and cut off the retreat of the two Japanese regiments to the north. The 
6th Company stayed where it was until 9 January 1944. 13 

Unknown to Sun and Stilwell the Chinese in the Kantau area faced a Jap- 
anese raiding force of about forty men which Tanaka had sent to swing wide 
round his left in order to cut behind the Chinese front-line positions and at- 
tack the truck parks, supply dumps, and command posts around Ningbyen and 
Shingbwiyang. The 18th Division had not expected to meet the Chinese. These 
latter posed so grave a threat to Tanaka's flank that their subsequent passivity 
puzzled their opponents as much as it gratified them. 14 

The fighting had by the end of 1943 cost the 38th Division 17 officers and 
298 enlisted men killed, and 20 officers and 409 enlisted men wounded. The 
bulk of the casualties were suffered by the 112th Regiment, and 356 replace- 
ments for it left Ledo on 3 January 1944. 15 

StilwelPs role had not been easy. Analyzing the operation in his usual 
fashion, he noted grave Chinese errors: "Dissipation of force. . . . Piece-meal 
action. . . . Extreme caution and extreme slowness of movement. . . . Fear of 
imaginary terrors. . . . Bad recon and security. . . . Fear of going around. . . . 

13 (1) Note cite d n. 8.1 Stilwell doubted Sun's aggressiveness. (2) NCAC History, I, 3lA. (3) 
G-3 Per Rpts, Hq 5303d (Prov) Combat Trs, 4 Tan, 11 Jan 44; Sketch of Sharaw, Hq Fwd Ech, 
5303d (Prov) Combat Trs, 28 Dec 43. NCAC Files, KCRC 

14 (1) Tanaka, Japanese Comments, Sec. II. (2) Japanese Officers' Comments, Incl 4, Tanaka. 

15 (1) Stilwell Sketch. Photostat 209, Stilwell Documents, Hoover Library. (2) Overlay and 
sketches, NCAC History, I, 30-32. (3) Dupuy Comments. (4) Overlay, Hq 5303d (Prov) Com- 
bat Trs, 28 Dec 43; Sketch, Disposition of Units of 1st Bn, 113th Regt as of 8 Jan 44, signed Capt 
N. R. Lester. Folder, Overlays, Albacore Opns, Sitreps Hist, etc., 11 Jan 44, NCAC Files, KCRC. 


right, front row, are Lt. Gen. Genzo Yanagida, Commander, 33d Division, General 
Tanaka, Commander, 18th Division, General Mutaguchi, Commander, Fifteenth Army, 
Lt. Gen. Sukezo Matsuyama, Commander, 56th Division, and Lt. Gen. Kotoku Sato, 
Commander, 31st Division. 

Result— Loss of men. Loss of chance to bag Japs." 16 He felt that the 3/ll4th 
was not pushing its envelopment, that it could have slipped up behind the 
Japanese from the south. Stilwell found it difficult to get an accurate idea of 
Japanese strength. One trusted staff officer thought them "awfully strong" 
across the Tarung; another, that there were only 400 Japanese facing the 38th 
Division. But his major problem was trying to make the Chinese more 
aggressive. 17 

The Opponents Shape Their Plans 

The warning of a Japanese offensive that had clouded Stilwell's enjoyment 
of Christmas was an exaggeration but not an error, for General Tanaka had had 
no intention of remaining on the defensive. He had planned to attack with his 

16 Handwritten note, 1943. JWS Misc Papers. 

17 (1) Stilwell Diary, 26-30 Dec 43. (2) Tanaka gives his strength in the engagement as 600 
men. Japanese Officers' Comments, Incl 4, Tanaka. 



55th and 56th Regiments, leaving the 114th Regiment to garrison Myitkyina, but 
had been overruled in mid- December 1943 by General Mutaguchi of the 15th 
Army. Since preparations were then actively under way for an attack on Imphal, 
Mutaguchi believed he could not spare additional motor transport units to give 
the necessary logistical support to the 18th Division. Tanaka therefore received 
orders that severely restricted him. The 18th's commander was told that any 
movement of reinforcements north of the Tanai must have Mutaguchi's per- 
sonal approval. The 18th could counterattack in the area around Maingkwan, 
near the south end of the Hukawng Valley; at the very least, it must hold 
Kamaing, just south of the ridge line that separated the Hukawng and Mo- 
gaung valleys. Weighing his orders, Tanaka began to plan with the hope that 
by the time the monsoon rains fell in May or June he would have created a 
stalemate somewhere near the ridge line. 18 

Therefore General Tanaka planned a delaying action down the Hukawng 
Valley, for which he had several advantages. The terrain was admirably suited 
to such an attempt, and the Japanese had complete and detailed knowledge of 
the Chinese order of battle. Moreover, the Japanese infantry were now sup- 
ported by the 18th Field Mountain Artillery Regiment with twelve 75-mm. 
mountain guns and four 150-mm. howitzers. Transport difficulties made it 
necessary to leave the remainder of the regiment in the rear. The engineers 
of the 18th Division were also present. After the war, Tanaka estimated his 
strength in the Hukawng Valley at 6,300 men. These plus the 114th Regiment 
stood between Stilwell and Myitkyina. 19 

Stilwell's plans immediately after the successful action at Yupbang Ga 
remained fluid. Probably he still feared that attack by the 18th and 31st Divisions 
of which he had been warned. Since his plans to meet that menace called for 
moving ahead to the line of the Tanai, and since General Sun was contemplat- 
ing an envelopment to trap the Japanese north of the Tanai, the operations 
immediately under way could go on as Stilwell shaped his next move and 
prodded Sun to move faster. To encourage the latter, and applying his bar- 
gaining technique of dealing with the Chinese, Stilwell told Sun on 3 January 
that the 3/1 12th would be released from army control if Sun took the little set- 
tlement of Taihpa Ga on the Tanai in two days. Looking farther ahead, Stil- 
well also asked General Wingate to move the U.S. long-range penetration 
group (Galahad) up to the Hukawng Valley. Wingate agreed. 20 

With the Generalissimo's refusal to cross the Salween, Galahad lost the 

18 Japanese Comments, Sec I. 

19 (1) Statement of Gen Tanaka, 5 Oct 51, Japanese Comments, Sec. I. (2) The Japanese 
strength estimate by Theodore H. White of 40,000 to 60,000 Japanese troops in the Hukawng 
and Mogaung valleys cannot be reconciled with what is known of the Japanese order of battle. 
The Stilwell Papers, p. 269. (3) Tanaka Interrog. OCMH. (4) Wkly G-2 Rpt, Hq 5303d (Prov) 
Combat Trs, 12 Jan 44. NCAC Files, KCRC. The 55th Regiment was on the south of a boundary 
line roughly Ningam Sakan-Yupbang Ga (both in the 55th Regiment sector); the 56th Regiment 
was on the north. Headquarters, 18th Division, was at Shingban. 

20 Stilwell Diary, 3 Jan 44. 



mission for which it had originally been intended and was available for others. 
On 6 January, General Merrill was assigned to command the regiment, vice 
Colonel Hunter, who became second in command. Galahad's designation 
reverted to unit, perhaps because it would have been incongruous to have a 
general officer command a regiment. 21 

Mountbatten offered British troops to spearhead the Chinese, but Stilwell 
preferred Galahad because both it and the Chinese had been trained in the 
American system of tactics. 22 Stil well's request, though later yielding major re- 
sults in combat, placed a logistic burden on the Advance Section and the com- 
bat zone for which no preparations had been made, because Galahad was to 
have operated under SEAC and Stilwell's decision reflected a last-minute in- 
spiration two months after the campaign began. There was no replacement or 
convalescent system for the American combat troops (since Galahad had 
been provided by the War Department for one mission of three months' dura- 
tion), while air support, both tactical and supply, had not previously been 
allocated with a view to support of U.S. infantry in north Burma. Arrange- 
ments by Stilwell's combat headquarters to support Galahad were therefore 
improvisations and many of them failed in combat. 23 

Seeking to give weight to the campaign from another direction, Stilwell 
sent a radio to Marshall asking that pressure be put on the Generalissimo to 
turn the Y-Force loose. Stilwell argued that the Generalissimo's inactivity 
would give to SEAC "the very reason they wanted to give up the attack on 
Burma." And he feared that if the Burma campaign was allowed to die away 
the Japanese would then be free to strike at Kunming either via Pao-shan or 
from Hanoi and French Indochina. 24 

Enveloping the Japanese Left Flank 

General Sun's plan, as he finally settled on it after many talks with Stilwell, 
called for a regimental combat team to continue the wide swing around 
Tanaka's left flank while the rest of Sun's division attacked Tanaka's front. The 
enveloping force, the 1 14th Regiment of the 38th Division plus the 6th Bat- 
tery, was sent circling south to cross the Tanai at Kantau while the 113th Regi- 
ment and the 2/1 12th in the center would wheel to the south, coming down 
from Yupbang Ga and reaching the Tanai, in the vicinity of Taihpa Ga. To the 
extreme north the l/112th would march eastward across the Tarung, then turn 
south. The 4th and 5th Batteries would support the 113th Regiment. Far to 
the west the 65th Regiment, 22d Division, was sent into the Taro Plain on a 
wide envelopment calculated to cut into the Hukawng Valley well behind the 

21 (1) Galahad, pp. 5-7. (2) Rad CHC 1241, Stilwell to Marshall, 3 Jul 44. SNF 131. 

22 Ltr, Mountbatten to Marshall, 16 Jan 44. Case 297, OPD 381 Security, A47-30. 

23 Notes by Gen Boatner on draft MS. (Hereafter, Boatner Notes.) OCMH. 

24 Rad, Stilwell to Marshall, 9 Jan 44. Folder, Rads, Stilwell to Marshall, Oct 43 to 25 Mar 
44, NCAC Files, KCRC. 



MAP 6 

18th Division, This wide swing was Stilweli's touch, the hallmark of a battle 
fought StilwelPs way. 25 

Attempting to envelop Tanaka's left, the 3/ll4th had crossed the Tanai at 
Kantau in the second week of December. It then encountered the Japanese 
along two small creeks, the Samp Hka and Mawngyang Hka. (Map 6) The 
jungle made it hard for the Chinese to organize; the supporting 6th Battery 

25 (1) "The 65th made a good river crossing," noted Stilwell in his diary. "... good dis- 
cipline and no grenade fishing." Its colonel, Fu, swore he would get after the Japanese. It was the 
"first time in his life," added Stilweli, "Fu ever saw a real envelopment." Stilwell Diary, 6 Jan 44, 
(2) G-3 Per Rpt, Hq 5303d (Prov) Combat Trs, 11 Jan 44. NCAC Files, KCRC. 



could not fire until fields of fire were cut, and while this was being done, in- 
filtrating parties of Japanese surrounded the battery. Fighting for their guns 
and their lives from 9 to 1 1 January, the Chinese cannoneers saved both. This 
Japanese attempt was greatly aided by the faulty disposition of the 3/ll4th, 
split into four fragments, no two of them in contact. Its commmander stayed 
in his rear command post. 26 

The regimental commander, Col. Li Hung, now arrived and quickly re- 
stored the situation. The several portions of the 3/ll4th began to attack, thrust 
the Japanese out from among themselves, and after the 2/ll4th arrived on 15 
January the revitalized 3d Battalion was able to force the Sanip crossing on 16 
January. This effort pushed a group of Japanese back into the first bend of the 
Tanai east of its meeting with the Tarung. The 2/ll4th was now on the right 
flank, where it faced perhaps one of the two Japanese companies present. 27 

Here the Chinese halted for a week, and when the l/ll4th arrived General 
Sun on 29 January sent the 3/ll4th back to rest and put the 1st Battalion to 
the slow job of digging the Japanese from their holes in the river bank. 

Stilwell thought this close to disobedience of orders, for he wanted Sun to 
move quickly and cut behind the 18th Division. On 13 January he had spoken 
very bluntly to Sun, asking him what orders he gave the ll4th Regiment and 
if there had been any word from Chungking to slow the operation. Stilwell 
pointed out that Sun's 38th Division had gotten weapons, supplies, medicine 
as no Chinese unit ever had before; Stilwell was going to bring up tanks, 4.2- 
inch mortars, U.S. infantry, and flame throwers, but before (underlined in Stil- 
well's draft) Stilwell put them in, he wanted to know whether Sun would obey 

Stilwell told Sun that if he (Stilwell) could not exercise the command that 
the Generalissimo had given him, he would resign and report the whole affair 
to the U.S. Government. "Regardless of what anyone else may say, I assure 
you that my report will be fully believed in Washington." General Sun was 
confronted with the prospect that Chinese lethargy in north Burma might 
mean withdrawal of all U.S. help from China. Stilwell, Sun was told, had been 
alone in his fight to convince the United States that the Chinese Army was 
worth helping. "If I am double-crossed by the people I am trying to help I am 
through for good and I will recommend very radical measures." Stilwell closed 
by saying that he had done his part; would Sun reciprocate? 28 But despite 
Stilwell's arguments and threats, General Sun did not meet Stilwell's ideas of 
how a dynamic field commander should conduct himself. 

For his part, Sun in late January told an American liaison officer with the 

26 (1) Dupuy Comments. (2) Stilwell Sketch. Photostat 215, Stilwell Documents, Hoover 
Library. (3) Overlay, Ln O, 38th Div, sub: Situation as of 8:00 PM, 16 Jan. Folder, Overlays, 
Albacore Opns, Sitreps, Hist, etc., 11 Jan 44, NCAC Files, KCRC. 

27 (1) Stilwell Sketch. Photostat 214, Stilwell Documents, Hoover Library. (2) Overlay cited 
n. 26(3). 

28 Memo, Stilwell for Sun at Ningam Sakan, Jan 44. SUP 7. 



38th that he had never had a failure in his life, and did not propose to have one 
now. His listener gathered that Sun would proceed with caution to avoid de- 
feat, rather than accept risks to gain a victory. 29 

The 114th Regiment plugged away at its own pace, taking care to leave no 
Japanese pockets behind, despite the obvious chance to clamp a roadblock 
across the road which led from Kamaing to the Japanese positions between 
the Tarung and the Tanai. 30 

At this point the 66th Regiment, 22d Division, was sent into action on the 
right of the 114th Regiment. Finding no enemy, it began to make its way 
toward the Kamaing Road. This action again opened the prospect of envelop- 
ing the Japanese. The regiment's move to the front had been delayed because 
of a missed ration drop, which had brought the columns to a standstill and 
greatly worried General Liao, who was having his first experiences with air 
supply in combat. 31 

The Capture of Taihpa Ga 

While the 114th Regiment was creeping through the jungle south of the 
Tanai, the 112th and 113th Regiments converged on Taihpa Ga from the north. 
The 1/1 13th crossed the Tarung near Yupbang Ga and sent patrols north to 
occupy Tabawng Ga on 13 January, then moved southeast to reach Kaduja Ga 
on the 15th. The 3/1 13th followed while the 2d Battalion stayed at Yupbang 
Ga in reserve. The regiment went southeast and lost two days "probing" at a 
Japanese delaying position. Then it bypassed on 17 January, left a company to 
contain the Japanese, and headed for Taihpa. Its pace was perhaps 150-200 
yards a day. Patrols from another unit, reconnoitering the banks of the Tarung 
to its junction with the Tanai, and then swinging east along the Tanai almost 
to Taihpa Ga, did not meet any Japanese in the area through which the 113th 
Regiment was moving so cautiously. 32 

The battered 112th Regiment swung wide to the north and east to Warang 
Ga, then halted and patrolled to the north, east, and south. The 113th Regi- 
ment at about this time (mid-January) extended its right until the 3/1 13th 
closed in on the Japanese positions on the Tanai bank. Clearing the Japanese 
from their footholds north of the Brangbram Hka and west of the Kamaing 
Road, on 21 January it reached Ningru Ga, less than a mile downstream from 
Taihpa Ga. 33 

29 (1) Dupuy Notes. (2) Ltr, Dupuy to Ward, 12 Sep 52. OCMH. 

30 Stilwell Sketches, NCAC History, I, 34A-34D. 

51 Rad Y-23, Boatner to Col John P. Willey, 12 Jan 44. Fwd Ech CP Rads Out, NCAC File, 

32 ( 1) Notes by Col Rothwell H. Brown on draft MS. (Hereafter, Brown Notes.) OCMH. (2) 
Ltr, with Incl, Brown to Ward, 25 Aug 51. 

33 (1) Stilwell Sketches. Photostats 212, 213, 218, 220-23, Stilwell Documents, Hoover 
Library. (2) Van Natta MS. (3) G-2, G-3 Overlays, 38th Div, 27 Jan 44; Overlay, 15 Jan 44, 
Albacore Opns, Sitreps; Rads 16-781, 21 Jan 44; Rads 14-77, 38th Div. NCAC Files, KCRC. 



GENERAL STILWELL AND GENERAL SUN are shown in conference at General 
Sun's headquarters. Col. Edward J. McNally, liaison officer with General Sun's troops, 
is seated at left. 

The 1/1 13th fought to within 1,500 yards of Taihpa Ga, then was ordered 
back to counter an expected Japanese flanking movement. This gave it the 
chance to clear a Japanese pocket previously contained along the Brangbram 
Hka. Artillery support, the 4th and 5th Batteries, was then brought up. The 
next 1,000 yards between the 1/1 13th and Taihpa Ga took two days to cross; 
then the 113th spent a week in what it called "preparations for attack." Gen- 
eral Sun's men were now approaching the Japanese strongpoint in the area, 
and Japanese resistance was more freely offered, with heavy shelling by 75- 
and 1 50-mm. pieces. 

At Taihpa Ga, the Kamaing Road crossed the Tanai on a long gravel bar 
which bullock carts could use quite well. At this point was the village itself, 
a humble collection of bashas, long since burnt out. About 800 yards upstream 
was a ferry across a stretch of fairly deep water. The Japanese used this ferry to 
bring supplies across the Tanai, rather than the vulnerable and easily spotted 
gravel bar. Here they had their strongpoint, well prepared and stubbornly 

The 3/1 13th was reinforced for the attack on the strongpoint. It moved 



through Taihpa Ga over 30 January-1 February and went on to attack the 
strongpoint. The 2/1 12th came back into action to remove a Japanese pocket 
on a line with, and three miles east of, Taihpa Ga, the advance bringing it up 
level with the 113th Regiment. Now back to full strength, the 112th Regiment 
gave security to the north, east, and west. The Japanese held stubbornly for 
several days and then quietly withdrew about 4 or 5 February. While the drive 
on Taihpa Ga had been under way, other Chinese units had been clearing the 
Japanese from the Taro Plain on Tanaka's left, and the obvious threat to his 
communications dictated his withdrawal, even though the Chinese did not 
emerge from the Taro Plain beyond his flank for another fortnight. When the 
Japanese yielded the north bank of the Tanai, a 4,500-foot airstrip was begun 
at Taihpa Ga, though Japanese shelling initially made the engineers abandon 
their work and their camp (to the delight of Chinese who promptly added to 
their stores of lend-lease equipment). 34 

Clearing the Taro Plain 

The Hukawng Valley is a corridor leading from north to south, and the 
Taro Plain has been compared above to a closet opening off the corridor, just 
inside the northern door. The Taro Plain is formed by the drainage system of 
the Hukawng Valley which tilts to the northwest and toward the Chindwin, as 
the Tanai is known in its southern stretches. About eight miles south of the 
Kantau ford, the hill line that forms the western boundary for the Hukawng 
Valley parts, and there the Tanai cuts a narrow, north-south gorge through the 
hills, a gorge that widens abruptly into the circular Taro Plain, with the village 
of Taro in its center. 

Given the mission in October 1943 of clearing the Taro Plain of Japanese 
so that the Ledo Road might safely pass to the east of it, the 3d Battalion, 112th 
Regiment, and its commander, Major Chen, had never left Ngajatzup on the 
extreme north edge of the plain. Stilwell's comment was: 

Sorry performance. Arrived about November 1. Sent one company forward. Pulled it 
back again. Thereafter did nothing. Maj. Ch'en cowered in dug-out. Terrific waste of am- 
munition. Told Sun to have him move or I would shoot or court-martial Maj. Ch'en. Sun 
sent [name illegible] to investigate. Ch'en killed by British grenade in his dug-out on 
December 27. (Report was during Japanese attack. There was none)." 

Unknown to Stilwell's headquarters, Tanaka was making a real effort to stop 
the unhappy Chen's battalion. When the 3d Battalion entered the Taro Plain, 
Tanaka rushed the 3d Battalion, 55th Regiment, to reinforce the Japanese 
"Pacification Unit" of eighty men stationed there. This battalion "made no 

34 (1) G-2, G-3 Overlays, 38th Div, 20 Feb 44; G-2. G-3 Overlays, Hq Fwd Ech (Prov) 
Combat Trs, 24 Feb 44; G-3 Per Rpts, 31 Jan 44, 8 Feb 44, 20 Feb 44. NCAC Files, KCRC. (2) 
NCAC History, I, 35-38. (3) Dupuy Notes. (4) Brown Notes. (5) Tanaka, Japanese Comments, 
Sec. II. 

35 Stilwell Sketch. Photostat 210, Stilwell Documents, Hoover Library. 



headway" so Tanaka then sent a Colonel Yamasaki and the headquarters of the 
53th, divisional units, and a second infantry battalion. Such a force was too 
much for the 3d Battalion, 112th, to handle. 36 

After Chen's death, the 3d Battalion was sent back to rejoin the 38th Divi- 
sion; and the 65th Regiment, 22d Division, plus attached Chinese engineers 
and the U.S. 46th Portable Surgical Hospital, was given the task of clearing the 
Taro Plain, with the extremely important added mission of cutting back east- 
ward into the Hukawng Valley to threaten Tanaka's flank. On reaching the 
exit of the Hukawng Valley, the 65th's commander, despite means more than 
ample for his mission, contracted the same lethargy that had hastened Chen's 
demise. Stilwell was in no mood to delay a few more weeks with Colonel Fu, 
and ordered Liao to relieve him. Stilwell's diary hints that his rhetoric rose to 
the occasion for he recorded: "Told Liao this included division commanders 
unless they watched their step. Also that Fu really should be shot. Liao took it 
OK though it shook him up." 37 

On 22 January General Sun was told about this affair, the news softened 
somewhat by presentation of a silk banner for the victory at Yupbang Ga, a 
dramatic contrast of the respective awards for lethargy and vigor. Having made 
his point, and hoping that he had given the Chinese a healthy shock, Stilwell 
restored Fu to his command by 26 January. Fu's later performance in combat 
was rated as excellent. 38 

The 65th Regiment moved forward immediately after Fu was relieved of 
command. Some U.S. observers believed that the 22d Division, from Liao 
down, looked upon itself as somewhat of a rival of the 38th, and wanted to 
show itself to advantage. The 65th's progress was aided by that of the 38th 
Division in the Hukawng Valley, for Tanaka was so concerned over his reverses 
there that he withdrew the major portion of the 55 th Regiment, leaving behind 
only the badly weakened and poorly supplied 3d Battalion of the 55th. These 
Japanese indulged in an ultimately fatal passivity which permitted the 65th to 
encircle them completely between the 23d and 25th of January. When the last 
shot was fired, the American liaison officers counted 323 enemy dead in a small 
area about halfway down the gorge through which the Tanai enters the Taro 
Plain. 39 

The next feature of importance was the Ahawk Hka, on whose far side was 
a trail, the Ahawk Trail, which was the shortest route from Taro back into the 

,6 Tanaka, Japanese Comments, Sec II. 

37 (1) Stilwell Diary, 20 Jan 44. The entry for the 19th makes it clear that Stilwell was angered 
because the 65th Regiment was still "stalled," the attack was a "fizzle," and Fu was "petering 
out." (2) Memo, Liao for Stilwell, 21 Jan 44, sub: Relief of Col Fu, 65th Regt. Item 266, Bk 3, 
JWS Personal File. 

38 (1) Stilwell Diary, 22, 26 Jan 44. (2) Boatner Notes. 

39 (1) Capt. Roy R. Van Dusen, Operations of the 2d Bn and 3d Bn, 65th Regt (2 2d Chinese 
Div), in the Battle for Taro, 29 Dec 1943-30 Jan 1944 (India-Burma Campaign). The Infantry 
School, Gen Sec, Mil History Committee, 1946-47. Captain Van Dusen's monograph has interest- 
ing tactical detail but says nothing of Fu's relief. (2) Notes by Col Van Natta on draft MS. (3) 
Tanaka, Japanese Comments, Sec. II. 



Hukawng Valley— and to a point behind Tanaka's flank. The Ahawk Hka was 
reached on the 26th and crossed on the 28th. The 3d Battalion of the 65 th put 
a block across the trail while the other two battalions went on to Taro. The 3d 
Battalion then went on down the trail toward the Hukawng Valley, pursuing 
the survivors of the Japanese Taro garrison, while its companions took Taro on 
30 January. 40 

After the war, Tanaka concluded that the loss of the Taro Plain meant "the 
failure of the whole division in its operations along the Tanai" and traced that 
loss in turn to his having had to withdraw the main strength of the 55th in an 
effort to redeem the situation around Kantau. 41 

The Allies Reorganize for the Next Effort 

With the arrival in the battle area of Liao's 22d Division and the American 
infantry of the Galahad force it was necessary to make appropriate changes in 
the headquarters directing this enlarged force. The changes began on 29 
January. Command of the service troops supporting the Chinese had been 
exercised heretofore by the 5303d Headquarters and Headquarters Company 
(Provisional) Combat Troops, whose members had been concurrently the staff 
of Chih Hui Pu. On the 29th the 5303d became an "area command" and on 1 
February 1944 the name was changed again to Northern Combat Area Com- 
mand, or NCAC as it was ever thereafter called. Use of the adjective combat was 
to insure its remaining under Stilwell, as by prior agreement line-of- 
communications areas in Burma were to be directly under SEAC. 42 

General Boatner was appointed Commanding General, NCAC, and kept 
his old post as Chief of Staff, Chih Hui Pu. Under NCAC were: . . such 
special and service units as may be placed in the Northern Combat Area Com- 
mand, except for SOS units specifically engaged in road construction and the 
auxiliary and service units necessary therefor. . . . Command of all combat 
troops in the Northern Combat Area Command remains as heretofore under 
the Commanding General, Chinese Army in India." 

As American, British, and Indian combat units entered north Burma, they 
were attached or assigned to NCAC, with the exception of the American long- 
range penetration group, Galahad, which on arrival was attached to the 
Chinese Army in India and later assigned to NCAC. 43 This brief initial attach- 
ment brought into focus some of the problems that arise when troops of 
different nationalities work together. In north Burma, some officers of each 

40 (1) NCAC History, I, 34A-34D, 36. (2) G-3 Per Rpts, Hq 5303d (Prov) Combat Trs, 
25, 26 Jan, 8, 15 Feb 44. NCAC Files, KCRC 

41 Tanaka, Japanese Comments, Sec. II. 

42 Boatner Notes. 

45 (1) History Ramgarh Training Center, 30 Jun 42-15 May 45, MS, p. 14. OCMH. (2) GOS 
11, 12, Hq USAF CBI, 29 Jan 44; GO 14, Hq Rr Ech USAF CBI, 5 Feb 44; GO 74, Hq USAF 
CBI, 17 Jul 44. (3) Memo, Stilwell for Wheeler and Boatner, 6 Apr 43. Item 214, Bk 3, JWS 
Personal File. 



force tended to feel that their men were being discriminated against in favor of 
the others. 44 The British headquarters in north Burma, Fort Hertz Area, under 
Brigadier J. F. Bowerman, was attached to NCAC on 1 February. 45 

CBI personnel entered NCAC's area at the Tirap River bridge, mile 5.75 on 
the Ledo Road. As of 2 December 1943 there had been no boundary between 
the communications and combat zones with the result that SOS in the forward 
zone was duplicating work of the troops under the 5303d in supporting the 
Chinese. Matters were arranged to keep Ledo, its installations, and the building 
of the Ledo Road under SOS while everything beyond the Tirap River went to 
NCAC. 46 

Three motives underlay the publication of the orders activating NCAC: 
to show the Chinese Government in writing that Stilwell was directly and 
formally commanding the Chinese Army in India; to show General Head- 
quarters (India) that the combat zone was well defined and commanded by 
Boatner and thus forestall any attempts to absorb it; to mark the zone of 
responsibility between SOS and combat troops, for it was considered "by all 
concerned that conditions make it impossible to establish the orthodox LOC 
Zone and Combat Zone." 47 To anticipate a later development, it may be 
observed that by April the principle had appeared of removing any possibility 
that the British and American combat units in north Burma might come under 
Gen. Cheng Tung-kuo, the newly appointed vice-commander of the Chinese 
Army in India, while on the other hand, it was desired to place absolute control 
of the pay and supplies of the Chinese Army in India in the hands of the 
Commanding General, NCAC. This was done by severing Ramgarh Training 
Center from NCAC. 48 As of 30 January there were 331 officers and 1,956 
enlisted men of the U.S. Army in NCAC, and nine officers and 240 civilians 
who were British subjects. 49 

Logistical Support 

The engineers, the medics, and the supply men followed close on the heels 
of the fighters. A complex line of communications, with airstrips, hospitals, 
supply points, motor shops, ordnance repair plants, and gas stations, began to 

44 (1) Boatner Notes. (2) For charges that Stilwell discriminated against the Chinese, see Ho 
Yung-chi, The Big Circle. (3) For charges that Stilwell discriminated against the Americans, see 
letter from Colonel Hunter to General Ward, 14 August 1951. OCMH. 

A "> Ltr, Hq 5303d (Prov) Combat Trs to Bowerman, 25 Jan 44, sub: Orders. AG 323.3, NCAC 
Files, KCRC. 

46 Ltr, Hq 5303d (Prov) Combat Trs to CG, Fwd Ech, USAF CBI, 2 Dec 43, sub: Responsi- 
bility of Comd and Opns Within Combat Zone. AG 323.3, NCAC Files, KCRC. 

47 Ltr, Hq 5303d (Prov) Combat Trs to CG, Hq Rr Ech USAF CBI, 20 Jan 44, sub: Organiza- 
tion, 5303d Area Comd. AG 323.3, NCAC Files, KCRC. 

48 ( 1) L ", Hq NCAC to Deputy Theater Comdr, USAF CBI, 13 Apr 44, sub: Responsibilities 
of CG, RTC, and CG, NCAC. AG 323.3, NCAC Files, KCRC. (2) Ramgarh Training Center pre- 
pared the Chinese Army in India for combat. The staff and instructors were American; logistical 
support was from the Government of India; unit administration, discipline, and replacements were 
the responsibilities of the Chinese Army. 

49 Strength Rpt, G-l Per Rpt, Hq 5303d (Prov) Combat Trs, 30 Jan 44. NCAC Files, KCRC. 



DRAINAGE CULVERTS are installed on the Ledo Road. 

twist its way south through the Hukawng Valley. The Red Cross clubmobile 
scurrying down the road, the ambulance plane and its tragic cargo, the bull- 
dozer driven by a Negro engineer masterfully thrusting aside the jungle, each 
at its appointed time passed the samurai blade rusting in the undergrowth. 

On 1 February 1944, the engineers began work on permanent road construc- 
tion in the Hukawng Valley, and on a combat road to support the Chinese 
divisions. The combat road, a hasty improvement of the existing Kamaing 
Road plus Kachin and Naga trails, ran through Shingbwiyang, Yupbang Ga, 
and Taihpa Ga, then south. The trace of the Ledo Road was moved to higher 
ground on the north. Forward construction units were rationed from combat 
supply points. 

In building the Ledo Road, location parties up ahead cleared a trace the 
width of a bulldozer and put in the center-line stakes. The final clearing 
averaged 150 feet. The route of the Ledo Road in some cases followed existing 
roads, a circumstance that did not greatly diminish the amount of clearing 
needed. Most clearing was by bulldozer. Combat trails and access roads were 
cleared to the necessary minimum that would permit heavy equipment to use 
them. In the valleys, the road was generally built on embankments in order to 
lift it above flood level. In mountainous regions, side-hill cuts were used. 



PIPELINE CROSSING A STREAM by means of cable suspension. 

The road itself had about seven culverts to the mile in the mountains and 
five to the mile in the lowlands. These culverts were most necessary as the road 
was a barrier to the normal runoff of water. Surfacing was with stream-bed 
gravel in the valley sections and, so far as hauling permitted, natural gravel in 
the mountainous sections. Surfacing was about ten inches thick on the average, 
and from twenty to twenty-eight feet wide. Compaction was by the normal 
road traffic. Two regiments of Chinese engineers did pioneer construction 
work. 50 

On 27 December 1943 Colonel Pick had opened a military road to 
Shingbwiyang and a convoy arrived that day. By 21 January 1944 General 
Covell saw more vehicles in Shingbwiyang than he could count. A subdepot 
was opened there in January with 21,600 cubic feet of storage space and 
facilities for bulk gasoline storage. An all-weather airstrip was built. Medical 
facilities were established. 

Parallel with the road ran the pipelines. Generally they were in the right of 
way and close to the road, but in some mountainous sections there were short 

50 (1) Rpt, Maj Gen Lewis A. Pick, CG, Hq Adv Sec USF IBT, to Lt Gen Raymond A. 
Wheeler, C G U SF IBT. 9 Aug 45 DP. 2 2-^. OCMH. (2) For background on the Ledo Road 
project, see|T7?7w7T7^1?jT?cT7oC%?«T1 



cuts using cable suspension over deep ravines and paths through terrain unsuit- 
able for road building. During the peak of construction on the pipelines 2,590 
men from engineer petroleum distribution companies and 2,750 general 
engineers were at work on them.'*' 

Medical reinforcements arrived in early 1944, the 42d, 43d, and 46th 
Portable Surgical Hospitals, the 13th Mountain Medical Battalion (less B and 
C Companies), and the 25th Field Hospital. The medical history of the 
campaign described the evacuation process of February 1944: 

Casualties thus passed through the hands of Chinese battalion and regimental detach- 
ments to surgical teams, thence either to the Seagrave Hospital on the Brambrang Hka or to 
the 25th Field Hospital at Ningam Sakan. Seagrave evacuated seriously-wounded to the 25th 
Field Hospital on about a 10-day evacuation plan. Patients requiring general hospital treat- 
ment or who would require more than six weeks of care were evacuated from the 25th Field 
Hospital to Ledo through the air clearing station at Shingbwiyang. 52 

Planning To Force a Decision 

Given the fact that Stil well's battle plans, as he strove with the 18th 
Division, always included either a double or single envelopment, it seems possi- 
ble to contend that Stilwell had in mind ending the campaign by a single 
decisive victory over the Japanese 18th Division. A successful envelopment 
would have been the speediest way of ending this part of the North Burma 
Campaign, and of offering the Japanese commanders their choice of yielding 
north Burma or canceling operations elsewhere in order to reinforce. 

Totaling his assets in January, Stilwell found he had immediately at hand 
the 38th Division, the 66th Regiment, 22d Division, and the Chinese 1st Pro- 
visional Tank Group (less the 2d Battalion), Col. Rothwell H. Brown, USA, 
commanding. 53 The Japanese he estimated as having a total of six divisions in 
Burma. He concluded that the 56th and 53th Divisions were tied down in the 
Salween and Arakan areas, respectively, that the 33d and 31st were contained 
by 4 Corps in the Imphal area, and that the 54th Division was protecting 
Rangoon and the Irrawaddy Delta against an amphibious attack. That left the 
18th Division for him to deal with. Since that division's 55th Regiment had been 
well worked over, it seemed unlikely the Japanese could muster more than five 
battalions to face his four regiments. 

51 (1) Ltr, Covell to Somervell, 21 Jan 44. Somervell File, Vol IV. Hq ASF. Ts of Opns, 
CBI 1944, A46-257. (2) Pick Rpt, pp. 15, 48, 51-58, 90-94, cite d n. 57)11 

" U.S. Army Med Service in Combat in India and Burma, 1942-45, MS by 1st Lt. James H. 
Stone, Medical Historian, IBT (hereafter, Stone MS), I, 122, 126, OCMH. 

53 The 1st Provisional Tank Group was then the only Chinese Army unit, under Stilwell, com- 
manded by an American. The group had an integrated Sino-American staff, a Chinese vice- 
commander (Col. Chao Chen-yu during most of the campaign), and an American medium tank 
platoon in support. The number of Americans, both officers and enlisted men, increased steadily 
from the initial component of 11 officers and 9 enlisted men, to a final peak strength of 29 officers 
and 222 enlisted men. Its equipment was American. See NCAC History, App. 5, First Prov Tank 
Gp, Chinese-American, 8 Aug 43-9 Mar 45. OCMH. 



The initial solution Stilwell reached took shape on 21 January. His decision 
then was to thrust an armored spearhead straight down the Kamaing Road, 
with a sizable infantry force following close behind. He expected an attack on 
this axis of advance to take his troops diagonally across the Japanese rear 
areas. 54 One uncontrollable factor Stilwell did not list in his estimate— weather. 
Dry ground would be essential. Orders were issued over the 21st and 22d. 

A few days passed, and it began to appear that the plan for an armored 
attack would have to be laid aside. January was all too plainly going to enter 
the records as a very rainy month, while Sun's attitude did not inspire Stilwell 
with confidence in his principal Chinese subordinate. A new plan would have 
to be made. 55 1 (Map 7)\ 

In his usual manner, Stilwell carefully charted his new course: 

February 4, 1944, it had become evident that the 38th Division could not be depended 
on for any further serious effort. The 112th had been scattered and cut up in November and 
replacements had arrived only at the end of January. The 114th, which had fought well, had 
lost about 60% of its company officers, and enlisted casualties had been fairly heavy. The 
113th was the only unit ready for further serious work, its advance to the Tanai having been 
slow and cautious. The division commander was showing a strong inclination to delay opera- 
tions interminably in the hope of avoiding more casualties; his failure to push the attack of 
the 114th was the cause of allowing the Japs an open route to the south which could easily 
have been closed. To operate seriously south of the Tanai it was necessary to establish a 
bridge-head, which the 38th Division commander obviously was reluctant to attempt. The 
plan had been to put the 1st Tank Battalion across the Tanai at night and jump off from the 
bridgehead at dawn, objective, Maingkwan. The l/66th was attached to the tanks and the 
113th was to follow down the road to take over successive positions. The 114th was to 
assemble at Taihpa in reserve and the 112th was to protect the flank east of the Tanai, 
advancing on Mashi Daru. 

The plan was altered on February 4 as follows: 22d Division, using 65th less one battalion 
(Taro garrison) and 66th less one battalion (attached to tanks) was to seize and hold line 
Yawngbang-Lakyen sending the 66th past the 114th, and the 65th over the Wantuk Bum. 
The 1 14th was to clean up south of the Tanai and the remainder of the 38th was to clear the 
area to the Tawang east of the Tanai. This was to be the first phase, followed as soon as 
possible by an advance to the south, using the tanks and the 65th and the 66th, in the hope 
of getting to Walawbum. By that time the 16th Brigade [of Wingate's Chindits] and the 
American brigade [Galahad] could make themselves felt at Lonkin and Shaduzup 
respectively. A plan was to be made to reduce Sumprabum and advance towards Nsopzup, 
and to edge towards Htawgaw from Luhow [illegible]. 

Bad weather had retarded Road work and hampered supply. The change in plan should 
give us time to build up a reserve in Shingbwiyang and Ningam, make progress on the 
Road, and build a field at Taihpa. It was felt that even though the seizure of a bridgehead 
might be delayed a few days, it would be a much better one and that the delay might well 
be made up for by cutting out the projected stop at Maingkwan and pushing on to the limit 
of our resources. A slower start, but a better organized one, with fresh troops eager to make 
good, and chance to go much further in less time, and with the added threats of the 
American and British LRPG's. 56 

54 (1) Stilwell Diary, 21 Jan 44. (2) Paper, sub: Estimate of Situation. JWS Misc Papers, 1944. 
(3) The Japanese 2d Division was also now in south Burma. Japanese Officers' Comments, p. 12. 
" Stilwell Diary, 2 1-30 Jan 44. 

"><> ( 1) JWS Misc Papers, 1944. (2) Stilwell Diary, 4 Feb. 44. 



"We can fight here instead of M[Maingkwan?] and can maybe get behind the 
Japs and clean up," wrote Stilwell in his diary for 4 February. So it was to be 
an attempt at a decision which if successful would mean that only Japanese 
stragglers and the 1 1 4th Regiment would stand between two Chinese divisions 
and Myitkyina. 

The first phase of the revised plan, the enveloping hook from the Taro Plain 
by the 22d Division, did not come off as planned. The 3/65 th of the 22d Divi- 
sion had been pursuing the Japanese from Taro down the Ahawk Trail. The 
2/65th was sent on from Taro to join it. 57 The rains were annoyingly heavy, but 
on the morning of 14 February Stilwell waited confidently for news that the 
65th Regiment had joined with the 66th Regiment, which had been operating 
in the Hukawng Valley, and had taken Yawngbang Ga, thus completing its 
part of the envelopment. In twenty-four hours it was very clear something had 
gone wrong. A Chinese officer of the 38th Division reported seeing the l/66th, 
which should have been five miles away from him. Stilwell spent all of 16 
February waiting for news of what had happened: "Is the 66th in Yawngbang 
or lost?" One set of American liaison officers reported the 66th Regiment was 
in place; another, that the regiment was miles away from its proper course. By 
noon of the 17th it was plain that the 66th Regiment had taken the wrong 
trail and lost its way. 

On the morning of 18 February Stilwell told General Liao that the 66th 
Regiment's performance had cost a chance to trap some Japanese, then went 
out personally to check on the regiment's location, with Liao accompanying 
him. The whole 66th plus the 3/65 th were in the neighborhood of the 66th's 
command post. The main trail, Yawngbang Ga to Lakyen Ga, was found and 
so was a captured Japanese document giving the Japanese withdrawal order. It 
later appeared that the 66th Regiment had taken a nameless village on 16 
February and thought it had taken Yawngbang Ga. The true Yawngbang Ga 
was occupied by the 65th and 66th jointly on 23 February. 58 

"If the Chinese 65th and 66th Infantry Regiments operating in the vicinity 
of Yawngbang had been prompt in closing in on our left rear flank on the 15 th 
or 16th, as predicted," wrote Tanaka in 1951, "the main force of the 18th 
Division would have faced a grave crisis." 59 

Summing up the results of the attempted first phase, Stilwell told Marshall 
that unseasonable rains and a mistake by the 66th Regiment, which had lost 
its way, cost a chance to catch some Japanese near Yawngbang. He felt that 
the 22d Division had done as well as the 38th Division and had been easier to 
command. Supply was improving with truck convoys coming down the Ledo 

^ G-3 Rpts, 29 Feb 44; G-2, G-3 Overlays, 25 Jan 44. Hq 5303d (Prov) Combat Trs. NCAC 
Files, KCRC. 

58 (1) Stilwell Diary, 9-18 Feb 44. (2) Overlays, Folder, X-Ray Force, NCAC Files, KCRC. 
,9 Tanaka, Japanese Comments, Sec. II. 



Road to Shingbwiyang and with transport aircraft using the Taihpa Ga airstrip. 
The Chinese enjoyed winning, and their morale was excellent. 60 

U.S. Infantry for the Second Phase 

Twice before Stilwell had tried to envelop the 18th Division, once along 
the south bend of the Tanai and again at Yawngbang Ga-Lakyen Ga. Both 
attempts had failed. Now, in accord with his plan of 4 February, he would try 
to put U.S. infantry across the Kamaing Road at Shaduzup, while the 22d and 
38th Divisions plus the armor pushed down from the north. 

When Galahad was released from SEACs operational control, Stilwell 
ordered Merrill to "close in on Ledo by 7 February," not an easy task, for the 
1,000-mile journey involved changes from a broad-gauge to a narrow-gauge 
railroad and from that to river steamer. The last echelon arrived at Margherita, 
near Ledo, on 9 February, and between 19 and 21 February the 5307th 
assembled at Ningbyen, near the front. Stilwell's foot cavalry was ready for its 
first mission. With it as far as Shingbwiyang went American newspapermen, 
who christened it "Merrill's Marauders." 61 

Stilwell's orders called for Galahad and the 113th Regiment 38th Divi- 
sion, to envelop the 18th Division's east flank and block the Kamaing Road near 
Shaduzup. The two units did not have a common commander and so lacked 
something of being a task force. After the war, Merrill recalled that he and Sun 
would have appreciated some sort of formal working arrangement but that 
Chih Hui Pu had limited them to co-operation. 

While the Americans and the 113th Regiment were cutting off the 18th 
Division in the Shaduzup area, the remainder of Stilwell's force— the other 
two regiments of Sun's 38th Division, the 22d Division, and the 1st Provi- 
sional Tank Group (-)— was to seize Maingkwan. Maingkwan had been a 
respectable little town before the war with permanent buildings and several 
thousand people, and Tanaka might elect to fight for it. So the attack on 
Maingkwan was set up as a smaller-scale edition of the complete operation, 
in that it was to be an envelopment, with the armor swinging around the 
Japanese right flank. Tanaka was to be offered unpleasant alternatives, fighting 
for the Maingkwan area at the risk of being surrounded, or fighting his way 
out through Galahad and the 113th Regiment, and thus giving up the whole 
of the Hukawng Valley. The remainder of the 38th Division, the 1st Provi- 
sional Tank Group (-), and the 22d Division were to push down from the 
north. If successful, the operation would pen the Japanese on the Maingkwan 
plain, and might well destroy the 18th Division, thus opening the way to 

For the operation, the boundary line between the 38th and 22d Divisions 

60 Rad 74, Stilwell to Marshall, 25 Feb 44. Folder, Rads, Stilwell to Marshall, Oct 43-25 Mar 
44, NCAC Files, KCRC. 

61 ( 1) The reference on page 276 of The Stilwell Papers is to the 16th Brigade of the 3d Indian 
Division, not to Galahad. Stilwell Diary, 9 Jan 44. (2) Galahad, pp. 1-8, 10, 16. 



was changed. The line now followed the Tanai to the Nambyu Hka, thence 
straight south to the Kamaing Road, giving the 38th Division the area to the 
north and east. 62 

Meanwhile, on the Japanese side there had been a growth of confidence. 
When the first shots were exchanged in October and November 1943, the 
Japanese had been almost contemptuous of the Chinese. Then in December 
they were sobered by contact with the material resources that the 22d and 38th 
Divisions commanded. But, as the fighting went on into February 1944, 
Tanaka several times saw a grave threat to his flank come to nothing because 
the Chinese moved so slowly. That the Chinese repeatedly moved as though 
to envelop, then let the opportunity slip because they did not exploit it, sug- 
gested to General Tanaka a way to defeat his powerful but lethargic opponents. 
Since Stilwell was spreading his forces wide in order to envelop, Tanaka conse- 
quently was operating on interior lines. He decided that "though threatened 
by enemy envelopment, we will exploit advantages of operations on interior 
lines, and, by utilizing every opportunity, defeat in detail the slow-moving 
Chinese forces without coordination on the exterior lines." 6J 

Thenceforth, Tanaka waited his opportunity, which would arise when 
Stilwell attempted his next envelopment. Tanaka's intentions and Stilwell's 
plans meshed perfectly; a major trial of strength was inevitable. The engage- 
ment resulting took place around the little settlement of Walawbum, a 
pathetic cluster of sagging uprights and fire scars where once a few Kachin 
families had pursued their simple, inoffensive lives. 

Walawbum is in the southern end of the Hukawng Valley. It is on the road 
running through the Hukawng Valley south to Kamaing. Ten miles or so south 
of Walawbum the traveler becomes aware that the ground is rising, for he is 
approaching the Jambu Bum, the ridge line forming the southern end of the 
Hukawng Valley. From Walawbum to the crest of the Jambu Bum it is about 
13.7 air miles. Seven miles to the west of the village site is the valley's western 
wall; the eastern boundary is from 10 to 14 miles away depending on the 
azimuth taken. 

The road to Kamaing, or Kamaing Road, takes a fairly straight course to 
the south until it reaches a point opposite, and about 10,000 yards west of, 
Walawbum. At this point it makes a 90-degree turn to the east, and runs 
almost due east until it reaches Walawbum. Perhaps a little more than halfway 
to Walawbum on this easterly course the road crosses the Nambyu Hka, at 
Kumnyen. When it reaches Walawbum, the road makes another 90-degree 
turn, this time to the south, and resumes its southward route to Kamaing. 

The visitor to Walawbum itself notices that as the Numpyek Hka passes to 
the east of the village, the ground to the east of the river is higher than on the 

62 ( 1) G-3 Per Rpt, Hq CAI, 29 Feb 44; G-3 Per Rpt, Hq 5303d (Prov) Combat Trs, 7 Mar 
44. NCAC Files, KCRC. (2) Notes by Gen Merrill on draft MS. (Hereafter, Merrill Notes.) 

63 Tanaka, Japanese Comments, Sec. II. 



west. Thus a force in position just east of Walawbum has the advantage of 
high ground and river moat in facing an enemy to the west, while its guns 
command the Kamaing Road. All about the area, save for a clearing at 
Walawbum, are the familiar trees and thick undergrowth, confining large units 
to the trails. 

Difficulties of terrain began to force alterations in Stilwell's plan even before 
it was fairly launched. The armored force was to have enveloped Maingkwan 
from the east but no suitable avenue of approach could be found so the tanks 
were ordered to move generally southeast on a course intended to clear 
Tanaka's eastern flank and permit them to join Merrill as an attached unit near 

The Operation Begins 

The Chinese- American tankers, supported by a battalion from the Chinese 
65th Regiment, got under way the morning of 3 March, from positions about 
5,000 yards north of Maingkwan. The first light contact with the Japanese was 
at 1300. By 1500 the tanks were about 5,000 yards northeast of Maingkwan, as 
they sought to flank to the east. In this area they received heavy Japanese fire, 
and Brown soon concluded that perhaps a regiment of Japanese faced him. 
Warning Stilwell of this potential threat to his (Stilwell's) left flank, Brown 
buttoned up for the night. During the darkness the Japanese made a number 
of attacks on the armor's perimeter, which were beaten off. Next morning, the 
Japanese were gone (for events near Walawbum were alarming them) and the 
tanks moved about three more miles southeast, which put them in line with 
Maingkwan. On 5 March, after a quiet night, they were ready to roll again, and 
to try to join Galahad near Walawbum. 64 

The progress of Brown's tanks meant that one part of the ring Stilwell was 
trying to clamp on the 18 th Division was moving into place. The adjacent 
segment of the ring's northern section, the 22d Division, was also moving 
ahead, and so far was in step with the tanks and Galahad. On 3 March, the 
64th Regiment was near Ngam Ga, east of the Kamaing Road. The 66th to the 
west was in contact with the third regiment, the 65th, as it emerged from the 
Taro Plain. The 66th bypassed Maingkwan on the west, moved south, then sent 
its 3d Battalion cutting back north to enter Maingkwan from the south. The 
Japanese yielded it; Tanaka was not going to waste his strength fighting for 
real estate. For all that, capture of Maingkwan by the 66th on 5 March gave 
that regiment the distinction of liberating the first major settlement in north 
Burma to be reoccupied by the Allies. After taking Maingkwan, the 66th 
moved south down the Kamaing Road, and with two battalions of the 64th 
Regiment, began attacking a fortified Japanese position in the Kumnyen area, 
to the west of Walawbum. The 65 th on 6 March was several miles southwest 
of Maingkwan, near the edge of the hills that mark the western boundary of 
the valley. Thus by the night of 5-6 March the 22d Division and Brown's 

64 Brown Notes. 




tanks were drawn up on a gently curving arc that ran through Maingkwan and 
tilted toward the south and east. 65 

Oral orders for the movement that was to put the rest of the trap in place 
were given Merrill on 22 February. He was ordered to cut the Japanese supply 
line, the Kamaing Road, well south of Walawbum, and to seek out and attack 
the 18th Division's command post, which was thought to be near Walawbum. 
Except for orders to avoid unnecessary heavy combat, Merrill had great freedom 
of action. At his disposal were the three battalions of Galahad, each in turn 
broken down into two combat teams. The teams bore the code names Red and 
White (1st Battalion), Blue and Green (2d Battalion), Khaki and Orange (3d 
Battalion). Each team included a rifle company, heavy weapons platoon, 
pioneer and demolition platoon, reconnaissance platoon, and medical detach- 
ment, with a combined strength of sixteen officers and 456 enlisted men. 

On receiving orders to move to his forward assembly area, Merrill sent his 
three intelligence and reconnaissance (I & R) platoons to check trails as far as 
the Tawang Hka, the first of the three considerable streams that crossed the 
line of march. At 0600, 24 February, the 5307th moved out, screened by the 
I & R platoons. Next day two of them clashed with Japanese patrols and the 

« (1) Rpts cite d n. 62(1*1 (2) Stilwell Diary, 5 Mar 44. 



point of one platoon, Pvt. Robert W. Landis of Youngstown, Ohio, was killed. 

The fortunes of war were with the Americans on this march, for radio 
communication between the 2d Battalion, 36th Infantry, and 18th Division head- 
quarters broke down at this time, so that the 18th was unaware that a 
semidetached American unit was operating off to its flank. 

On the afternoon of 28 February a liaison aircraft brought orders to shift 
the roadblock site to Walawbum itself. When the 5307th crossed the Tanai, 
last of the three river barriers, it went into an assembly area on 2 March, and 
Merrill called his staff and commanders together to get Chih Hui Pu's orders. 
The orders called for cutting the road on either side of Walawbum, the 2d 
Battalion (Colonel McGee) to the west, the 3d Battalion (Colonel Beach) to 
the south, and the 1st Battalion (Colonel Osborne) to patrol along the 
Nambyu Hka north of the Kamaing Road. Positions near Walawbum would 
be held until the 38th Division relieved the 5307th. Merrill's own plans had 
been to put his battalions at the Nambyu Hka, but Stilwell now believed 
Brown's tanks could reach and hold the river line. 66 

All battalions were away by dawn of 3 March. Patrols clashed with the 
enemy throughout the day, and the 3d Battalion had a sharp fight at Lagang 
Ga, killing thirty Japanese in seizing the area needed for the building of a drop 
field. One of the battalion's two combat teams, Khaki, stayed at Lagang Ga to 
build and protect the dropping zone. Orange Combat Team kept on to the high 
ground east of Walawbum and dug in, its heavy weapons commanding the 
Kamaing Road. During the day, 3 March, the 1st and 2d Battalions were still 
occupied in moving to their assigned positions. On 4 March the 2d Battalion 
put itself across the Kamaing Road west of Wesu Ga, making a roadblock. 
"This makes the net fairly good . . . ," wrote Stilwell. 

On 1 March General Tanaka had been told that the Americans were at 
Walawbum. 67 It was premature, but this was the chance Tanaka had awaited. 
Quickly analyzing his situation, he decided that the Chinese 22d and 38th 
Divisions were moving so slowly that he could contain them with a small rear 
guard while the main strength of the 18th Division hurled itself on the Ameri- 
cans. On 2 March he made his decision, and the movement back began on 
3 March. The Americans had cut behind the 33th Regiment on Tanaka's east in 
order to place themselves across the Kamaing Road, so the 33th in turn was to 
hit them on their northern flank while the 36th drove for the place where the 
Kamaing Road crosses the Nambyu Hka. 68 

66 (1) Galahad, pp. 16-20. (2) Tanaka, Japanese Comments, Sec. II. (3) Ltr, Merrill to 
Sunderland, 10 Apr 48; Ltr, Merrill to Ward, 26 May 52. OCMH. 

67 (1) Galahad, pp. 21, 23. (2) Memo, Hunter for Kent R. Greenfield, Chief Historian, 
OCMH, 11 May 45, sub: Comments on "Merrill's Marauders." OCMH. (3) Stilwell Diary, 4 
Mar 44. 

68 (1) Lt. Gen. Shinichi Tanaka, The Fighting of the Japanese 18th Division in the Kamaing 
and Walawbum (Burma) Area, MS, May 1949- (Hereafter, Tanaka Narrative.) OCMH. (2) 
Tanaka, Japanese Comments, Sec. II. 



GALAHAD's Fight at Walawbum 

Small parties of Japanese blundered into the American lines on 3 March, 
but there were no organized attacks until the 4th. (Map 8 J The drop field at 
Lagang Ga was attacked at dawn on the 4th but the garrison held. Orange 
Combat Team opened the battle in its sector with mortar fire on Walawbum, 
drawing mortar and 75-mm. fire in return from the 56th Infantry Regiment, 



which had been Tanaka's left flank, as it assembled for attack. When it moved, 
the 36th tried to cross the river and work around the Americans' flanks, where 
it met booby traps and ambushes thoughtfully prepared for just such a contin- 
gency. Some Japanese did cross to the east, but this failed to affect the course 
of the action, and seventy-five dead Japanese were counted, as against one 
American dead and seven wounded. These latter were evacuated by air the same 
evening. 69 The 2d Battalion, 5307th, to the west, received much heavier blows, 
for it was closer to the main Japanese concentration. Beginning late on the 
evening of 4 March the Japanese attacks lasted until the next day. 70 Ammuni- 
tion began to run low in the 2d Battalion, and during the last thirty-six hours 
of the fight it was without food and water. 

Orange Combat Team of the 3d Battalion was under no great pressure 
during the day of the 5 th, but the men believed Japanese reinforcements were 
being brought up from Kamaing for an attempt to remove the roadblock. The 
Americans from Stilwell on down were convinced they had trapped Tanaka; it 
did not occur to any of them that Tanaka was trying to destroy the Americans. 
Several times during the 5 th the Japanese appeared to be forming for an attack, 
but mortar fire seemed to be successful in breaking up such attempts. To the 
south, Allied aircraft could be seen bombing and machine-gunning what 
Orange Combat Team took to be Japanese reinforcements. One indication that 
the Japanese were increasing their strength in the immediate area lay in their 
being able to force Orange Combat Team's I & R Platoon back across the river 
about noon of the 5 th. 

Meanwhile, to the north and east of where the 2d Battalion and Orange 
Combat Team were fighting, heavy and constant pressure from Brown's 1st 
Provisional Tank Group was forcing Tanaka to alter his plans. Brown's recon- 
naissance had found a good trail running south from Tsamat Ga, and on the 
morning of the 5th the tanks moved out through the jungle. After the engi- 
neers had prepared a small stream for crossing, the tanks broke into a freshly 
evacuated Japanese bivouac area. Jungle vines looping across the trail from 
either side, and connecting masses of vegetation and trees, made effective 
obstacles as they slowed down the tanks by catching their turrets; not until late 
afternoon did the armor break out on the trail running east and west between 
Maingkwan and Wesu Ga. Almost immediately the tanks encountered what 
seemed to be a company of Japanese defending a small but marshy stream. The 
stream did not seem fordable, so Brown attacked by fire alone. Unknown to 
him, his tanks were firing on Tanaka's division headquarters, and now lay 
squarely between the 18th 's headquarters and its 36th Regiment. 

Further compounding the Japanese commander's problems, Brown had 
brought his tanks down the trail that the 33th Regiment was to have used for its 

69 Galahad, pp. 21-22. 

70 (1) Tanaka Narrative. (2) Tanaka, Japanese Comments, Sec. II. General Tanaka claims 
the 3/56tb captured the Nambyu river crossing with one rapid attack on 4 March, the one 
major point at which his account differs from the American. 



attack on Galahad. The 55th as a result had displaced westward so that 
instead of coming up on line with the 56th for an assault on Galahad it had 
in a sense fallen in behind the 56th in a sort of column formation. The 56th 
itself was making little progress against Galahad. 

Tanaka decided to give up his attempt to crush Galahad. Instead, he 
decided to swing his force around the American roadblock, using a Japanese- 
built trail of whose existence the Americans were unaware, and then to 
re-establish his front facing north in an east-west line across the Kamaing Road 
once more. So resolved, Tanaka, or one of his staff", late on the 5th of March, 
picked up a field telephone and began to give the necessary orders. The bypass 
road over which the 18th was to withdraw had been built by the Japanese 
engineers some days before: 

The Engineer Regiment commander, Colonel Fukayama, had considered the possibility 
of reversals in our position and, in order to facilitate the withdrawal of the division, he had 
previously cleared a secret jungle trail about 20 kilometers long leading from the vicinity of 
Lalawng Ga to Jambu Hkintang on his own initiative. This trail was used in the withdrawal 
of the main body of the division. 

Meanwhile, within the 2d Battalion's roadblock, Sgt. Roy H. Matsumoto 
had been monitoring Japanese telephone conversations, for the 18th's wire 
communications passed through the roadblock and had been tapped by the 
Americans. Sergeant Matsumoto immediately picked up Tanaka's order, and in 
a matter of minutes the American commanders knew of Tanaka's intentions. 

In the light of this disclosure, and since the 2d Battalion lacked food and 
water, the decision was reached to withdraw the 2d at nightfall. The 2d was 
ordered to fall back on Wesu Ga, receive fresh supplies there, then march south 
to join the 3d Battalion below Walawbum. Colonel McGee and his men 
arrived at Wesu Ga at noon of 6 March, and went on to rejoin the 3d. 71 

Late on the next day, the 6th, the fighting reached its climax, as Tanaka 
sought to move his division from between the 22d Division on the north of his 
current positions and the American roadblock to the south. Supported by the 
other combat team of its battalion, Orange Combat Team spent the morning 
and early afternoon of the 6th in bracing itself for the expected supreme 
Japanese effort. Shells from Japanese medium artillery suggested it might not 
be long in coming. The pack animals suffered severely from this preparation, 
but the men in their foxholes with overhead cover took few casualties. At 1715 
an estimated two companies of Japanese in line of skirmishers, with heavy 
supporting fire, crossed the river. The American mortars continued their work; 
the automatic weapons held back until the Japanese were within fifty yards. 
Two heavy machine guns, which had a clear field of fire along the river bank, 
were especially effective. The Japanese failed, leaving many dead on the open 
ground east of the river and on the river banks. Orange Combat Team found 

71 (1) Galahad, pp. 24, 27-28. (2) Tanaka Narrative, p. 2. (3) Tanaka, Japanese Comments, 
Sec. II. (4) Brown Notes, p. 6. 



its ammunition low, and so sent a request for more to Khaki Combat Team, 
which was about one hundred yards downstream. 72 

From the Japanese point of view, withdrawal was begun just in time, for 
the leading elements of the 113th Regiment, Chinese 38th Division, which 
had been teamed with Galahad, met General Merrill about a mile northeast 
of Walawbum at 1600 hours, 6 March. The arrival of the Chinese meant that 
the outcome of the battle could not be an Allied defeat; the problem was now 
to make the outcome a real and if possible a decisive victory. 

There was an unfortunate incident when the remainder of the 38th Division 
and an American patrol met near Walawbum on 7 March. The recognition 
signal had been arranged as three bursts of three rounds each. This necessarily 
meant there would be firing when the Chinese and Americans met. When the 
tops of the American helmets, which looked not unlike the Japanese pot 
helmet when their brims were invisible, appeared through and over the brush, 
there was a brief exchange of fire in which three Chinese were wounded before 
identity was established. 

As soon as the Chinese were present in strength, Merrill by arrangement 
with Sun but over the protests of his battalion commanders withdrew 
Galahad on the morning of 7 March. His intent was to circle south and cut 
the Kamaing Road again farther to the south. Merrill was very mindful of 
StilwelPs order to keep his casualties down, and since he had held his position 
until the 38th Division appeared, he believed he had complied with Stilwell's 
orders. In effect, his decision removed Galahad from the battle of 
Walawbum. 73 In announcing his withdrawal from the Walawbum area, 
Merrill told Colonel Brown, the tank commander, that he was relieved of 
attachment to Merrill's headquarters and should radio Stilwell for further 
orders. 74 

Apparently determined to make every moment count, Brown while waiting 
his next orders attempted on the morning of 7 March to arrange a co-ordinated 
tank-infantry attack toward Walawbum. He asked the commander of the 
Chinese 113th Regiment to join him in a thrust direct at Walawbum, but the 
Chinese officer was unable to agree because his orders were to hold at Wesu 
Ga, about 4,000 yards north-northeast of Walawbum. Brown returned to his 
own bivouac area and talked the situation over with staff members of his 
attached Chinese infantry battalion and his own people. While the discussion 
was under way a battalion commander of the 64th Regiment, Chinese 22d 

72 Galahad, pp. 27-28. 

73 (1) Merrill's Marauders. (2) The statement by Gen. Dr. Ho Yung-chi in The Big Circle, 
page 86, that the 5307th broke and ran at Walawbum is denied by Hunter and Merrill. Some doubt 
is cast on Dr. Ho's charge by his own claim that the 38th Division subsequently chased the 
Japanese "to the walls of Walawbum." Walawbum was half a dozen burned-out grass shacks 
on a dirt road. On page 44 of Merrill's Marauders is a picture of the Chinese moving in while 
the Americans moved out, which event may have given rise to the story. (3) The brief account 
in Fred Eldridge's Wrath In Burma (New York: Doubleday Doran & Company, 1946), page 
221, is not accurate. (4) Ltr to Sunderland, cited n. 66(3). (5) Merrill Notes. 

74 Ltr, Brown to Ward, 25 Aug 51. OCMH. 



Division, appeared. Placed between two Chinese divisions which because of 
communications difficulties were operating almost independently, Brown had 
to make his own arrangements for a suitable role, so he turned to the newly 
arrived battalion commander with his suggestion for an attack. This officer, as 
it turned out, had no specific orders from his own higher headquarters, but he 
did have an aggressive disposition, and promptly adopted Brown's suggestion 
that tanks and infantry join to move south from their present site and place a 
block across the Kamaing Road as it runs due east toward Walawbum. 

About 1500 on 7 March, the tanks and the battalion of the 22d Division 
placed themselves across the Kamaing Road in not one but two places, respec- 
tively one and two miles west of the Nambyu Hka. The infantry set up a 
roadblock, while the tanks moved out aggressively along the road to east and 

One tank company moved west along the Kamaing Road and had bad 
luck, for it met nothing but an impassible stream that halted further progress. 
A second company went east along the Kamaing Road until it came to a bridge 
that proved to be well covered by Japanese antitank guns. The lead tank was 
almost across the bridge when it gave way, dumping the luckless vehicle into 
the water. Antitank fire ripped through the thinly armored portions and killed 
all but one of the crew. 

The third tank company was the most successful. Its commander decided 
to turn off the Kamaing Road onto a trail that showed signs of heavy traffic. 
This may have been Tanaka's evacuation route, for the tanks encountered a 
body of Japanese on the march and scattered them. Two weeks later a mass 
grave of 200 dead Japanese was found in the area. On the evening of the 7th 
the tanks reassembled after dark. 75 

The "Big Squeeze Play" 

When dawn broke on the 8th, Stilwell's forces, with the already-noted 
exception of Galahad, occupied areas forming a great arc whose several seg- 
ments were seeking out the Japanese near them, the task of search no easy one 
in the country around Walawbum. In order from west to east were two regi- 
ments of the 22d Division, the tanks, and the 113th Regiment of the 38th. The 
22d's regiments were not arrayed neatly in line but had their elements over a 
considerable area which included the ford over the Nambyu Hka at Kumnyen— 
thus explaining how Brown's tankers met a battalion of the 22d near the 
Nambyu Hka. Tanaka seems to have made his withdrawal from the area 
through which elements of the 2 2d were moving but they did not keep pressure 
on him, and to this he later attributed his escape from the Allied arc that might 
have become a deadly ring: "The cautious movement of the Chinese forces 

75 Brown Notes, pp. 7-11. 



engaged in the frontal attack made it possible for the 18th Division to save 
itself." 76 

Back at his headquarters, Stilwell on 8 March checked over the situation as 
it appeared to him on the basis of the information available at his headquarters. 
He believed Merrill to be still in place along the Kamaing Road south of 
Walawbum, Sun's 38th Division to be on the field, Brown's tanks to be in 
action, Liao's 22d Division to be coming down from the north, and the 18th 
Division to be withdrawing. Victory on a major scale appeared to be in sight, 
so Stilwell issued orders for a "big squeeze play" of every unit he had to con- 
verge on the 18th Division and crush it. The 64th Regiment Stilwell thought to 
be on the west of the Japanese position so the 64th was to attack towards the 
east; Sun's 113th Regiment was to attack towards the southwest, and so on 
round the arc. 77 What followed illustrates how dependent the commander is 
on forces and factors beyond his immediate control or even knowledge, for the 
shortcomings of the radio net made Stilwell and Merrill unaware of each other's 
intentions and movements. Moreover, the several units, Chinese and American, 
often did not know of each other's locations and maneuvers, and so instead of 
a co-ordinated assault on an encircled 18th Division what actually took place 
was a battalion and regimental commanders' battle as units engaged what 
Japanese they could find. 78 

In compliance with Stilwell's orders, the tanks and the 38th Division began 
the moves that ultimately placed them in what was left of the village of 
Walawbum. The tanks, which were on the Kamaing Road west of Walawbum 
and separated from it by the Nambyu Hka, moved out eastward to seek a ford. 
Japanese antitank fire covering the ford on the main road at a place called 
Kumnyen Ga discouraged thoughts of crossing there so bulldozers covered by 
infantry prepared a new crossing one and a half miles to the north. Once this 
was done, Brown was ready to take his tanks across the stream and attack 
Walawbum itself. The site itself had nothing of value but since the Japanese 
had put up such a fight in the immediate vicinity Brown felt they must value the 
location and so he resolved to make a determined effort to get it. To do this, 
he decided to send two tank companies across the newly prepared ford to bypass 
the Japanese position covering the old ford at Kumnyen Ga, and go on to 
Walawbum, while one company of tanks made a frontal attack on the 
Japanese. 79 

The envelopment went very smoothly, and as soon as the two tank 
companies had cut behind the Japanese positions and were on the main road 
they continued on toward Walawbum. Unfortunately, the tank company and 

76 (1) NCAC History, I, 63. (2) Unidentified overlay, showing opns, 1-3 Mar 44. Folder, 
X-Ray Force, NCAC Files, KCRC. (3) Tanaka Narrative, p. 2. (4) Quotation from Tanaka, 
Japanese Comments, Sec. II. 

77 Stilwell Diary, 8 Mar 44. 

78 For comments on the communication problem, see Merrill Notes. 

79 Stilwell's diary entry of 9 March describes Brown as having done his part. 



A CHINESE COOK makes his way forward with his cooking equipment. 

Chinese infantry which were making the frontal attack moved only as far as 
Kumnyen Ga itself, which lay on the west of the ford. It was then 1600 and 
the Chinese infantry refused to move farther. The appearance of Chinese 
soldier cooks with rice kettles suggests one reason for the delay. But the 
Chinese tankers were willing to keep on, and the tanks crossed the Nambyu 
Hka at the main road. They soon rejoined the other two companies in 
Walawbum. At dark, the tanks pulled back out of Walawbum to the Nambyu 
Hka, where the Chinese infantry had halted, as Brown did not want the armor 
to be without infantry support during the night. Next morning, the 9th, the 
tanks and their accompanying battalion moved back to Walawbum and found 
it, as Brown later recalled, "swarming with people from the 64th Infantry and 
the 113th Infantry" who, to the tankers' great disgust, disputed the latter's 
claim to have taken Walawbum. 80 

The largest share of the 38th Division's work at Walawbum had fallen to 
the 2d and 3d Battalions of the 113th Regiment. It was the 2d Battalion that 
had established contact with Galahad and relieved that unit. The 1st Battal- 
ion, which followed it, was kept in reserve. The 3d Battalion, which had 

80 Brown Notes, pp. 8-11. 



moved up by forced marches, re-established the roadblock west of Walawbum 
that Galahad's 2d Battalion had been forced to yield several days before, and 
kept its hold during the 8th and 9th after considerable righting. Presumably it 
was this battalion the tanks met in Walawbum on the 9th. General Sun used 
the 112th Regiment to guard his rear and eastern flank lest the Japanese attack 
him from across the hills which mark the eastern boundary of the Hukawng 
Valley, the Kumon Range. The 114th, Sun kept in reserve. 81 

After Galahad was withdrawn from Walawbum and on its way to cut the 
Kamaing Road again farther south, communications between Merrill and 
Stilwell were finally re-established late on the evening of the 8th. Merrill was 
ordered to halt, for Stilwell was now looking toward the next phase. Stilwell 
recorded in his diary that he learned of Galahad's maneuver only after issuing 
orders for a co-ordinated action by the 64th, 113th, Galahad, tanks, artillery, 
and the rest. Weighing the situation, Stilwell decided that his orders to Merrill 
had not been clear enough. In saying, "use your discretion" he had meant to 
keep casualties down, not "go roaming." Stilwell's conclusions and his willing- 
ness to assume responsibility for not making his orders more clear deserve 
respect, but the communications difficulties that kept Stilwell in the dark as to 
the movements and location of the several units, plus the extreme caution of 
the Chinese 22d Division, seem major factors in Tanaka's successful withdrawal 
from the Walawbum area. The 18th Division made good its escape from 
Stilwell's trap, but in so doing it had yielded control of the greater part of the 
Hukawng Valley to the Allies, and the Chinese Army in India could celebrate 
a well-earned victory. 82 

Between the Hukawng Valley and the Mogaung River valley is the ridge 
barrier of the Jambu Bum. Once Stilwell was fairly over the Jambu Bum, he 
would be in a corridor, the Mogaung valley, which leads directly into the 
Irrawaddy valley and Burma proper. The North Burma Campaign was begin- 
ning to yield results. For them, the 22d and 38th Divisions were paying a toll 
in casualties. From fall 1943 to 18 March 1944, the campaign had cost 802 
Chinese dead and 1,479 wounded, plus 530 undifferentiated casualties. Of the 
dead, 539 were from the 38th Division. 

Medical aid for the Chinese victors of Walawbum was provided by the U.S. 
Army. The 42d, 43d, and 46th Portable Surgical Hospitals worked with the 
22d Division. Surgical teams from the Seagrave Hospital Unit, the 25th Field 
Hospital, and the 13th Mountain Medical Battalion were with the 38th Divi- 
sion. An ambulance shuttle up the Kamaing Road evacuated patients to the 
25th Field Hospital at the new Taihpa Ga Airstrip. 83 

81 G-3 Per Rpts, Chih Hui Pu, 7, 12 Mar 44. NCAC Files, KCRC. 

82 Stilwell Diary, 8, 9 Mar 44. 

83 (1) NCAC History, App. 5, History of 1st Prov Tank Gp, Chinese- American, 8 Aug 
43-9 Mar 45, pp. 8, 14. (2) Rad Y-285, 26 Mar 44; Rads, Fwd Ech OUT. NCAC Files, 
KCRC. (3) Stilwell Diary, 4 Mar 44. (4) Stone MS, pp. 131, 133, 135. 




The two-month stalemate around Yupbang Ga in November-December 
1943 finally obliged Stilwell to leave CBI Theater headquarters and assume 
command in the field. The arrival of reinforcements that gave the Chinese a 
considerable numerical superiority, the constant air supply that gave them mo- 
bility unprecedented even in jungle warfare as the Japanese conducted it, 
enabled Stilwell, Sun, and Liao to make a series of envelopments that forced 
Tanaka to loosen his grip on the strategic area where the Tanai and the Tarung 
join. The long stalemate was broken and a war of movement began. Stilwell 
decided to use a fresh regiment of U.S. infantry for his next envelopment, 
while Tanaka decided that because his opponent moved so slowly he could 
safely mass his forces against the attempted envelopment and defeat his op- 
ponents in detail. A confused and stumbling clash around Walawbum fol- 
lowed, in which the Japanese were outfought and thrust to the south. This 
was the first series of Allied victories in the North Burma Campaign and a 
major slice of Burma was freed from the Japanese. But four months had gone 
by since the first shots were fired. Progress forward had been slow, if victorious. 


The Burma Campaign 
in the Balance 

When the soldiers of the Chinese Army in India entered Walawbum, deep 
in Japanese-held Burma, they were for all practical purposes masters of the 
Hukawng Valley and well on the way to Myitkyina. But their success did not 
guarantee enthusiastic support of the Burma campaign by either SEAC or the 
Chinese Government, while the Japanese at one point were within a few miles 
of making the whole North Burma Campaign impossible. So far was SEAC 
from favoring a campaign in north Burma, that Mountbatten sent a mission to 
London and Washington to plead for cancellation of the campaign, while the 
Americans in north Burma believed that on his part the Generalissimo was 
ordering Generals Sun and Liao to go very slowly indeed. For some weeks the 
campaign hung in the balance, while Stilwell did what he could to tip that 
balance on the side of vigorous action to break the long blockade of China. 

The SEXTANT Decisions Challenged 

On 4 March 1944, when the fighting at Walawbum was moving to its 
peak, Admiral Mountbatten visited Stilwell at the latter's invitation. The rela- 
tion between the Supreme Allied Commander and his acting deputy was in 
need of adjustment because of a chain of events stretching back to the preced- 
ing January. 1 The Generalissimo's unwillingness to cross the Salween and ad- 
vance into Burma, plus the long-felt desire of Mountbatten's commanders in 
chief to bypass Burma and to postpone major operations until after the defeat 
of Germany, led SEAC's planners to revive, and Mountbatten to approve, the 
old plan for an attack against the Netherlands Indies that had first been pre- 
sented in May 1943. SEAC's staff argued that the clearing of north Burma to 
break the blockade of China was out of step with global strategy because of the 
now quickened pace of operations in the Pacific. They recommended that the 
Combined Chiefs of Staff order SEAC to keep its resources out of the Burma 
fighting and conserve them for a major offensive in the fall of 1944 or the 
spring of 1945. The first phase of this operation should be the securing of a 

1 The Stilwell Papers, p. 282. 



foothold in Sumatra. Success in the East Indies should be exploited into the 
South China Sea, with the goal of opening a port on the China coast. Ger- 
many's fall was expected to release the means for this. 2 

On learning of this development, the deputy CBI Theater commander, 
Maj. Gen. Daniel I. Sultan, had promptly warned Stilwell of it. Telling Stil- 
well that Mountbatten was about to send a mission to Washington and Lon- 
don to present this plan, General Sultan had suggested that Stilwell counter 
with a mission of his own. Stilwell was, as acting Deputy Supreme Allied 
Commander, subordinate to Mountbatten; he was also Chief of Staff, China 
Theater, with obligations to China and Generalissimo Chiang Kai-shek. A de- 
cision to postpone any effort to relieve China until after Germany had been 
defeated might have the gravest repercussions in China. As Chief of Staff, 
China Theater, Stilwell therefore had decided to send his own mission to 
Washington: Generals Boatner and Ferris, Colonels Francis Hill and Francis 
G. Brink. 3 Moreover, Stilwell was an officer in the service of the United States, 
and there lay his first loyalty. As commanding general of a U.S. Army theater 
of operations he had been charged with several missions, such as support of 
the airline to China, which might be adversely affected by the strategy SEAC 
was proposing. He also knew that the objections of the Joint Chiefs of Staff 
to bypassing Burma had been placed on the record at the Washington con- 
ference of May 1943. Unhappily, he failed to inform Mountbatten of his de- 
cision to send the mission. 

As soon as the situation around Taihpa Ga permitted, Stilwell had placed 
his arguments directly before SEAC. At a Supreme Commander's meeting on 
31 January, Stilwell stated flatly that the entire concept of attacking Sumatra 
and bypassing Burma was wrong. While he fully realized the advantages of 
opening a Chinese port that could ultimately result from the proposed oper- 
ation, Stilwell believed this could be accomplished by an overland advance of 
ten Chinese divisions in about one month, after the Ledo Road to supply them 
had been opened. Stilwell believed that the road would be opened sooner than 
the SEAC War Staff anticipated and wanted to know if account had been taken 
of the existing road to Myitkyina, the Kamaing Road, from which he was then 
only twenty miles away. As for the argument that a change in strategy was 
necessary because of the faster schedule of Pacific operations, Stilwell did not 
see how that made it necessary to wait six months to fight elsewhere instead 
of fighting the Japanese at once in Burma. 4 

2 (1) Rad AM 37, Sultan to Marshall, Stilwell, and Hearn, 6 Jan 44. Item 1602, Bk 5, JWS 
Personal File. (2) CM-IN 5998, Sultan to Marshall, 9 Jan 44. (3) Rad, Mountbatten to COS and 
Prime Minister, 8 Jan 44. SEAC War Diary. 

3 (1) Rad AMMDEL AG 13, Stilwell to Marshall, McNarney, and Handy, 2 Jan 44; Ltr and 
Incl, Hearn to Mme. Chiang, 11 Jan 44. Item 263, Bk 3, JWS Personal File. (2) Rad RE 76, 
Sultan to Stilwell, 13 Jan 44. Item 16, Bk 6A, JWS Personal File. 

4 Min, SAC's 5 2d Mtg, 31 Jan 44. SEAC War Diary. Stilwell probably meant the motorable 
portion of the Kamaing Road. 



StilwelPs representations made no impression, for a clash of national policy 
was involved far deeper than differing views on strategy. The United States 
wanted to support China; the British Commonwealth, to liberate its prewar 
territories in Burma and Malaya, and again raise the British flag over Singa- 
pore. In retrospect, British emotions and British policies should be fully under- 
standable in terms of similar emotions and policies that resulted in the Amer- 
ican effort to liberate the Philippines. As Admiral Mountbatten stated the 
problem in his report: 

The Americans, on the one hand, regarded Assam and Burma primarily as part of the air 
and land line to China; the British, on the other hand, saw the liberation of Burma as an 
end in itself, and as a step on the road to Malaya and Singapore. It seemed necessary to 
establish the fact that, although the two conceptions were divergent in motive, in execution 
they were complementary and inseparable. 5 

To the Prime Minister, viewing the war from the lonely summit of his posi- 
tion, all of Burma seemed far from Japan, and he feared that if the British 
forces were mired in Burma they would not have their just share of victory in 
the Far East. He "wished, on the contrary, to contain the Japanese in Burma, 
and break into or through the great arc of islands forming the outer fringe of 
the Dutch East Indies." As he wrote his memoirs several years later, Churchill 
added: "It is against this permanent background of geography, limited re- 
sources, and clash of policies that the story of the campaign should be read." 6 

Unknown to Stilwell, the subject of a new directive for SEAC was an 
urgent topic of conversation among War Department planners. Their conclu- 
sions, a reaffirmation of their May 1943 views, were reached before StilwelPs 
mission to Washington could present his reactions to Culverin (SEAC's plan 
for an attack on Sumatra). On 12 January the Asiatic Section, Operations Divi- 
sion, examined the proposed attack on Sumatra in response to Sultan's warn- 
ing on the 6th. The section pointed out that President Roosevelt's 20 Decem- 
ber 1943 message to the Generalissimo strongly indicated a land route to 
China. Furthermore, as Japanese strength in Burma increased, the enemy's of- 
fensive capabilities were correspondingly improved, a factor to be weighed in 
any consideration of operations against Sumatra. 7 

From north Burma Stilwell had been reporting successes against the Jap- 
anese and asking when 4 Corps would move forward from Manipur State to 
engage them. In response, the Joint Chiefs of Staff asked the British Chiefs of 
Staff to inquire of SEAC when it proposed to advance the 4 Corps, and what 
its mission would be. 8 

5 Mountbatten Report, Pt. B, par. 2. 

6 Winston S. Churchill, Closing the Ring (Boston: Houghton Mifflin Co., 1951), p. 561. 

7 (1) Memo, Col William H. Wood, Actg Chief, Asiatic Sec OPD, for Chief, S&P Gp OPD, 
12 Jan 44, sub: Opns, SE Asia Theater. Case 228, OPD 381 Security, A47-30. (2) See p. 79, 
above; pp. 297-98, below. 

8 (1) CM-IN 857, 2 Jan 44; Rad MS 37, 22 Jan 44; CM-IN 14766, 23 Jan 44. These radios 
were "Action Handy; Info Marshall, McNarney." (2) CCS 452/5, 25 Jan 44, sub: Opns in Burma 



The JCS, disturbed by the evident SEAC tendency to postpone operations 
in Southeast Asia, and approving the Joint Staff Planners' view that the cap- 
ture of Myitkyina in the current dry season was essential to the U.S. policy of 
developing China into a base capable of supporting Pacific operations, warned 
the CCS of their apprehension that further delay in issuing a CCS directive to 
SEAC would result in very little being done before the monsoon began. The 
Joint Chiefs went on to say that every means available for operations in north 
Burma should be employed immediately. They recommended that SEAC be 
ordered to seize and hold Myitkyina before the 1944 rains and that 4 Corps be 
ordered to cross the Chindwin and move on to central Burma. To this view, 
the British Chiefs of Staff replied that they did not agree with the proposed 
directive, but that they were not prepared to advance counterproposals, pending 
the outcome of conversations with Mountbatten's mission. 9 

The British Chiefs of Staff reaffirmed a SEAC assertion that SEAC was 
making the maximum offensive effort that its line of communications could 
support. They doubted whether the 4 Corps could seize or hold the Shwebo- 
Monywa area before the monsoon and whether the Ledo Force could advance 
to Myitkyina in time to bring the Ledo Road forward before the monsoon. In 
the opinions of the British Chiefs, if this could not be accomplished Myitkyina 
could not be held. 

Aware of these exchanges between the Joint and British Chiefs of Staff, 
Admiral Mountbatten complained that the views of the Joint Chiefs were 
based on Stilwell's representations, with all that implied of insubordination by 
the latter. He accused Stilwell of inspiring a press campaign against Culverin 
and asked that Stilwell's relief from SEAC duty be arranged. General Marshall 
quickly pointed out to Field Marshal Sir John Dill, Chief of the British Joint 
Staff Mission in Washington, that the JCS paper had been prepared before 
Stilwell's staff officers had arrived in Washington and before the JCS had seen 
Stilwell's objections to Culverin. Fully consistent with the previous course of 
U.S. service opinion the JCS paper had stemmed from the OPD memorandum 
of 12 January, and the latter was drafted before General Sultan suggested 
Stilwell send his mission. 10 Therefore, Stilwell's mission on its arrival had no 
need to present his views, for they were identical with the positions already 
taken by the Joint Chiefs of Staff and President Roosevelt. 

At this point Stilwell received concrete indication that the President's 
attitude toward him was becoming more favorable. From Washington, General 
Boatner radioed that the President revealed complete sympathy with Stilwell's 

» (1) Memo, Wood for Chief, S&P Gp OPD, 9 Feb 44, sub: Opns, SEAC; Rpt (JCS 678/1), 
JPS, 16 Feb 44, sub: Opns in SE Asia, 1943-44. ABC 384 (Burma) 8-25-42, Sec 5, A48-224. (2) 
CCS 452/6, 17 Feb 44, sub: Opns in SE Asia; CCS 452/7, 20 Feb 44. (3) CM-OUT 8897, 21 
Feb 44. Case 297, OPD 381 Security, A47-30. 

10 ( 1 ) CCS 45 2/8, 26 Feb 44, sub: Opns in SE Asia. ( 2 ) Rad SEACOS 99, 2 1 Feb 44. Item 
66, OPD Exec 10. (3) Ltr, Marshall to Dill, 23 Feb 44. Case 297, OPD 381 Security, A47-30. (4) 
Ltr. Mountbatten to Wedemeyer, 20 Feb 44. SEAC War Diary. (5) Stilwell's Mission to China, 
|Ch. IX. | 



efforts to advance in Burma. Not since the spring of 1942 had Stilwell received 
any evidence that the Commander in Chief was supporting him. Unimpressed, 
Stilwell replied to Boatner that he would trade the sympathy for one U.S. 
division. 11 

Reporting on his interview with the President, Boatner stated that Roose- 
velt said he was more dissatisfied with the progress of the war in Burma than 
anywhere else. The two men discussed the role of the British and Chinese 
ground forces in Burma. Boatner urged the President to aid in persuading 
the British and Chinese to take a more aggressive role in Burma. The Presi- 
dent agreed to address the Prime Minister and urge immediate aggressive 
action in compliance with the Sextant decisions, but he said nothing about 
similar representations to the Generalissimo. 12 

Lending point to Boatner's arguments was a widely circulated OPD estimate 
of the strength of the contending parties in Burma. OPD believed that there 
were in India 1,654,094 British and Indian troops, 44,036 U.S.-trained Chinese, 
and 2,943 U.S. infantry, or 1,701,073 men, against a G-2 and SEAC estimate of 
150,000 Japanese in Burma. 13 The strength estimates made by OPD are, how- 
ever, not a true picture. The Japanese strength in Burma, exclusive of naval 
and air personnel, was approximately 252,000. 14 As for the Indian forces, OPD 
did not take into account the very considerable forces needed to restrain the 
warlike and turbulent tribes of the North- West Frontier Province, the need of 
garrisons to maintain internal security in an empire of some 400,000,000 
peoples of diverse races, and the administrative establishments required for 
support of the Indian divisions in the Mediterranean and Middle East, as well 
as in Burma. 

Colonel Hill of Stilwell's party underscored the relationship between taking 
Myitkyina and increasing Hump tonnage above its current plateau of a 
13, 000-14,000- ton level. Hump tonnage was limited by the high density of air 
traffic at the Assam gateway and by the lack of instrument let-down facilities 
in China. As a result the Hump was rather closely confined to 10,000 tons a 
month in bad weather, 12,500 tons in good. The ATC, said Hill, was suggest- 
ing that a more southerly route by way of Myitkyina would have to be flown 
to lessen traffic density. 15 However, so long as Japanese fighters could use the 
Myitkyina airstrip, ATC's transports would have to stay well to the north. 

The President, as he had promised Boatner, told Churchill that he was 
expecting Stilwell to take Myitkyina by the end of the dry season, and that 
he thought Stilwell could hold there if 4 Corps did its part. Roosevelt ex- 
pressed grave concern over recent trends in SEAC's strategic thinking, which 
favored Sumatra and Malayan operations. He concluded that advantageous as 

11 Rad RE 296, Sultan to Stilwell, 20 Feb 44. Item 44, Bk 6A, JWS Personal File. 

12 Rpt, Boatner, 18 Fe b 44. on interv with Roosevelt. Item 245, Bk 15, OPD Exec 9. 
11 App. A to Rpt rire dn Q ml 

14 Japanese Officers' Comments on third CBI MS. 

15 Min, Washington Conf, 9 Feb 44. ABC (Burma) 384, 8-25-42, Sec 5, A48-224. 



a successful Culverin might be, more could be gained by an all-out drive into 
upper Burma to take Myitkyina and thus increase the airlift to China. This 
would permit building air strength in China and insure essential support for 
the projected cross- Pacific advance to the Formosa-China-Luzon area. Churchill 
replied that SEAC would not withdraw or withhold any forces from the cam- 
paign in upper Burma for the sake of Culverin or any other amphibious 
operation. This reply was reassuring, for General Sultan had heard gossip at 
SEAC headquarters that troops would not be committed to current operations 
lest such delay Culverin. 16 

The Japanese Create More Command Problems 

While the lengthy radioed exchanges on the new directive for SEAC were 
taking place, the Japanese seized the initiative in Burma, thus justifying the 
apprehensions so often expressed by General Marshall. On 7 January, Imperial 
General Headquarters had finally approved the major portion of the plan to 
attack India so long urged by Burma Area Army, and had directed: 

In order to defend BURMA, the Commander-in-Chief of the Southern Army may occupy 
and secure the vital areas of Northeastern india in the vicinity of imphal by defeating the 
enemy in that area at the opportune time. 17 

Significant of the limited scope of the intended operation, which was given the 
simple code name U, this directive was issued in accordance with an Imperial 
General Headquarters basic order of summer 1942 which had given a defensive 
role to the Japanese garrison of Burma. 18 

In the opinion of the former chief of the Operations Section, Imperial General 
Headquarters Army Department, another order by that body would have had to 
be issued for an advance into Assam Province beyond the Imphal area. All 
things considered, so ambitious an operation by 1 5 th Army did not seem 
possible to the Japanese command. Some members of the 15th Army command, 
including General Mutaguchi himself, did want to descend into Assam, and 
expected that such a bold offensive would be permitted by Tokyo when the U 
operation had been successful. 19 

16 (1) Memo, Marshall for Leahy, 24 Feb 44, with Incl, sub: Proposed Msg from President to 
Prime Minister. Case 297, OPD 381 Security, A47-30. (2) Rad, President to Prime Minister on 
SEAC Strategy, 25 Feb 44. SEAC War Diary. (3) Rad 592, Prime Minister to President, 2 5 Feb 
44. Folder, Prime Minister-President, OPD Exec 16. (4) The President's radio is quoted in 
Churchill, Closing the Ring, page 561. 

17 (1) Imperial General Headquarters Army Directive 1776, 7 Jan 44 (hereafter, IGH Army 

Dir ), GHQ, Far East Comd, Mil Hist Div, Imperial General Headquarters Army Orders. 

OCMH. (2) SEATIC Bull 240, 9 Jul 46, pp. 3-5; SEATIC Bull 247, 22 Apr 47, pp. 3-4; Lt Gen 
Kitsuji Ayabe in Bull 247, p. 6; SEATIC Bull Special Int Bull, Pt. V, P- 22. MID Library. (3) 
Japanese Study 89. 

18 ( 1) IGH Army Order 650, 29 Tun 42. (2) For a discussion of Japanes e strategy in South- 
east Asia in the summer of 1942, see \$tilwell's Mission to China, Chapter V.| (3) Japanese Com- 
ments, Sec. I. 

19 Statement, Col Hattori, 5 Oct 51; Statement, Col Fujiwara, 5 Oct 51. Japanese Comments, 
Sec. I. 



R. t/ahn.3 iaie 

MAP 9 

These details were unknown to the Allied staffs and commanders, but on 
4 February 1944 they saw the beginnings of the crucial stages of Operation U 
when the 33th Division attacked in the Arakan to lure SEAC's reserves away 
from Imphal. 20 (Map 9) 

The Japanese attack was not a surprise to SEAC, to CBI headquarters, or to 
General Headquarters (India), though Stilwell, like Marshall, had always 
feared an attack on Kunming. As early as 15 July 1943, Lt. Gen. Geoffrey 
Scoones, 4 Corps commander, had predicted the Japanese move. In December 
1943 the likelihood of such an attack was under active discussion in SEAC. 
Wingate was advised of its probability on 16 January 1944. SEAC's joint 
intelligence accepted the view that a Japanese offensive was imminent and 
objected to a War Staff estimate which forecast a rather passive Japanese 

20 Mountbatten Report, Pt. B, par. 68. 



By 8 February 1944 Stilwell's mission to Washington could report that an 
estimate of the situation by General Headquarters (India) suggested a major 
Japanese offensive was about to begin. Almost immediately, CBI staff officers 
began to study the implications of such an attack in terms of the air supply 
needs of surrounded Allied units. On 14 February Admiral Mountbatten told 
the British Chiefs of Staff that a Japanese attack on 4 Corps was planned to 
follow a successful offensive in the Arakan. On 5 March he inquired of General 
Giffard, the senior British Army commander in SEAC, how that officer 
proposed to reinforce Imphal to meet the expected Japanese attack. 21 

The Japanese 28th Army in the Arakan had three divisions, the 2d, the 54th, 
and the 55th. Convinced that SEAC contemplated an amphibious descent on 
the Arakan coast, the Japanese employed the 2d and 54th Divisions in the 
profitless tasks of building coastal defenses and keeping a sharp watch for 
hostile landing craft, as well as the more useful work of forwarding supplies 
to the veteran 55th Division, victors of the Battle of the Sittang Bridge, Feb- 
ruary 1942. The 55th Division's mission was to place the Indian divisions facing 
it in such peril as to attract British reserves from the decisive point at Imphal. 

The Japanese striking force made its way around 15 Corps' east flank with- 
out discovery and attacked on 4 February. So quickly and hard did the Japanese 
strike that on 6 February they overran the headquarters of the 7th Indian 
Division. Similar Japanese tactics had yielded very considerable results in 
previous years. Here was a threat. 

But General Slim, commanding Fourteenth Army, the headquarters 
immediately controlling 4 and 15 Corps, saw the opportunity presented by the 
Japanese move. Exploiting the advantages of air transport and air supply he 
quickly massed a force that far outnumbered the Japanese by bringing up the 
26th Indian Division and moving the British 36th Division into a supporting 
position. A third division of two brigades, the 81st West African, was operating 
in the Kaladan valley, parallel to the Arakan coast. 

To meet the Japanese attack, the three brigades of the 7th Indian Division, 
the brigade of the 5th Indian Division attached to it, and the division service 
troops formed perimeters and went on air supply while the rest of 15 Corps 
drove toward them. Called "baby tortoise" or "beehive" tactics by the baffled 
Japanese, this type of operation was something new in Burma fighting. The 
unit training, and equipment, of the Indian Army had greatly improved in the 
past years. By cutting the Indians' supply lines, the Japanese had expected to 
force them back in confusion. Instead, though the Japanese were carrying out 

21 (1) Rad 213, Stilwell to Bennett, 7 Dec 43; Rad, Stilwell to Marshall, 9 Jan 44. Folder, 
Rads, Stilwell to Marshall, Oct 43-25 Mar 44, NCAC Files, KCRC. (2) CM-IN 2094, Stilwell to 
Marshall, 4 Jan 44. (3) Note, Secy Plans on War Staff, 1944, p. 12; Note, Gen Wingate-Situ- 
ation in North Burma. Items 4, 16, SEAC Info Bk, OCMH. (4) Min, 1st Conf with Offs USAF 
CBI, 8 Feb 44. ABC 384 (Burma) Sec 5, 8-25-42, A48-224. (5) Rad RE 240, Evans to Stilwell 
and Sultan, 11 Feb 44; Rad RE 260, Evans to Stilwell, 14 Feb 44. Items 34, 38, Bk 6A, JWS Per- 
sonal File. (6) Minute SC4/438/G, SAC for Giffard, 16 Mar 44; SAC Dir to CinC, 11 Army Gp, 
5 Mar 44. SEAC War Diary. (7) British Comments, Serial 56. 



their mission of containing superior forces, they themselves, depending on 
what supplies could trickle to them over jungle trails, were effectively encircled 
and were running out of supplies in a few days. 22 In two weeks' time, 7th 
Indian Division restored communications with its service elements and held 
the initiative. 

That these new Allied tactics involved air supply posed new problems for 
Stilwell. On 13 January 1944, well before the Japanese struck, Stilwell's staff 
warned him that the Arakan situation (where an African division was soon to 
require air supply) might result in a request for diversion of Hump aircraft 
and Hump tonnage at any time. A week after the battle began such a request 
came from SEAC. The Arakan supply commitments strained Mountbatten's 
resources to the limit, and as it would soon be time to fly Wingate's new 
and ambitious long-range penetration group effort into Burma, Mountbatten 
wanted to divert thirty-eight aircraft from the Hump. He wanted Stilwell's 
concurrence before he radioed the CCS, but no reply came before the radio was 

Stilwell's reply was a refusal for he had understood that only one Japanese 
battalion was causing the trouble. How, he asked, could he justify a diversion 
on those grounds to the Generalissimo? The CCS, however, agreed, providing 
no more than thirty C-47's were taken off the Hump. Mountbatten could argue 
that the spirit of the SEXTANT discussions with the Generalissimo contem- 
plated Hump diversions in an emergency, and this was one. 23 Hump tonnage 
fell from 14,431 tons in Febr uary to 10,934 tons in March, then rose to the 

mid-13,000's in April and May. (See Chart 5.) 

When the Arakan crisis ended in late February with the driving of the 
Japanese from their key roadblocks, both sides could find cause for satisfaction. 
Churchill proclaimed: "The enemy has been challenged and beaten in jungle 
war-fare. His boastfulness has received a salutary exposure." 24 From all indica- 
tions, the Japanese task force took very heavy casualties. But the men who 
planned to meet again at Yasukuni Shrine expected this, and Burma Area Army 
believed the sacrifices of its men had in fact mired GifTard's and Slim's reserves 
in the Arakan while the decisive action was being fought out hundreds of 
miles to the north. The Japanese had in effect sacrificed the 55th Division and 
greatly weakened their defense of the Arakan, so much so as to make its future 
loss inevitable. This meant the Japanese would have to win at Imphal to 
make the Arakan sacrifice worth while. They did not perceive that Slim and 
Giffard, by the rapid movement of their strategic reserves, had taken the first 
major step toward defeating them in detail, for they did not realize that Slim 
and Giffard could shuttle whole divisions by air. 

22 (1) Japanese Study 89. (2) Mountbatten Report, Pt. B, pars. 60-92. 

23 (1) Rad RE 248, Evans to Stilwell, 12 Jan 44; Rad RE 260, Evans to Stilwell, 14 Feb 44; 
Rad, Stilwell to Evans. Items 15, 38, 47, Bk 6A, JWS Personal File. (2) Rad SEACOS, 18 Feb 
44; Rad JSM toWar Cabinet Office, 25 Feb 44. SEAC War Diary. 

24 Owen, The Campaign in Burma, p. 65. 



The principal concern of the Japanese staffs after the Arakan battle was the 
slowness with which their 15th Division was moving into place for the attack 
on Imphal. The drive on Imphal was to have begun on 22 or 23 February. The 
Japanese D Day came and went and the 15th Division was still not ready to 
move. Anxiously the Japanese staffs watched the time they had bought by 
sacrificing the 55th Division slip through their fingers. 2 ^ 

Mountbatten and Stilwell Meet 

On 2 March General Marshall told Stilwell to see Mountbatten at once and 
re-establish cordial personal relations with him. Marshall referred to some press 
indiscretions, which Mountbatten thought had been StilwelPs work, and told 
Stilwell that he knew the latter bore no responsibility whatever for them, that 
they originated with a U.S. naval officer. Nevertheless, the Prime Minister and 
British Chiefs of Staff had at once concluded that Stilwell was responsible for 
them. Their irritation had been increased by the mission Stilwell sent to 
Washington without telling Mountbatten. Though Marshall had assured Dill 
and Mountbatten that the Joint Chiefs had reached their views independently 
of Stilwell, harm had been done. Hinting at great combined operations to 
come, which made Allied unity essential, Marshall asked Stilwell to close the 
rapidly widening breach between himself and Mountbatten. 26 

As was his custom, Stilwell carefully drafted what he wanted to tell 

Mission to Washington was to inform War Department of my plans and arguments, 
report on situation, and attempt to put China in proper perspective. I am CofS for CKS. 
I command CBI Theater. I have a mission. My orders re SEAC are to use U.S. forces to best 
purpose and I have done so. I am serving under Slim. I diverted Hump tonnage. I am 
following orders. I have never been appointed as Deputy [Supreme Allied Commander, 
SEAC]. Opinions differ. I have stated mine frankly. E.G., 'Ledo Road cannot be built or 
held, nor can it carry any tonnage.' It is being built. It can be held— not of course if the 
British leave the whole job to two Chinese divisions. It will carry tonnage. It is now carrying 
20,000 tons a month without trying. 

Burma can be taken. The Japs can be licked. (Slim's job. Our job.) The road can be 
opened. All we need is pressure on CKS, which I have struggled to get. The problem in 
front of us is Burma. That is what I am working on. 

What are the objections to me? I have assured you that once a decision is made, I will 
play ball. Am I supposed to sink my opinion before that time? And help to put over a point 
of view I do not believe in? Why should I take the opinions of your staff and not my own? 

Have I covered up to you on the Chinese? Have I withheld possible help? Do you 
believe now what your experts said of the [Assam] LOC? Have I squawked about my status? 
Have I talked for publication? (Indian Press). Have I meddled in politics? (Mr. Churchill 
told me last May he understood my position and wanted to help). 

25 Japanese Comments, Sec. I. 

26 Rad WAR 4654, Marshall to Stilwell, 2 Mar 44. Item 2062, Bk 6, JWS Personal File. 



Trident, Quadrant, and Sextant all bore down on opening up China. Why should 
I be panned for trying to do it? 

Have I mentioned publicity? (Show him the Road article.) Does that tend to create 
suspicion or not? (Constant belittling [in the Indian and British press] of American effort 
here with the Chinese). . . . 

Army Commander did not have any remarks on military effort. I have no desire to mix 
in political matters. 

What is it? Do you want me to resign? I have never been appointed, so that is easy. Do 
you want me to be relieved entirely from this Theater? Or do you want the CBI abolished? 
( Have you political reasons for your attitude?) You asked me to be frank. I should like you 
also to be frank. I can stand it. 27 

The meeting between Mountbatten and Stilwell passed off pleasantly and 
successfully. Stilwell kept no notes on what he said but reported to Marshall 
that he "ate crow" because he had not told Mountbatten of his mission. "We 
are good personal friends," he went on, "and our relations have never been 
stiff." Stilwell stated that he was carrying out Mountbatten's orders to the best 
of his ability and would carry out future orders as soon as the decision was 
made. At Cairo, he went on, he had tried to get a statement of the United 
States China policy but this had not been vouchsafed him. Since this guidance 
was lacking, he was trying to protect U.S. interests, carry out his original 
mission, and comply with Trident, Quadrant, and Sextant decisions. 28 

Admiral Mountbatten was pleased by the meeting and told Dill of Stilwell's 
apology. Mountbatten's letter to Dill suggests a certain reserve but he consid- 
ered that he and Stilwell had settled their current problems and had effected 
a personal rapprochement. However, Mountbatten added: 

He really is a grand old warrior but only the Trinity could carry out his duties which 
require him to be in Delhi, Chungking and the Ledo Front simultaneously, and I still think 
Al. Wedemeyer or Sultan should be appointed as Commanding General for the American 
SEA theater and that Stilwell's command should be confined to China though he could 
certainly continue with the title of deputy SAC, SEA since he had never really done 
anything about those duties during the whole time I have been out here. 29 

Later events showed that Stilwell was willing to focus his attention on China, 
to be relieved of a post that was of purely symbolic significance, and to let 
others cope with problems of administration and command south of the Hump. 
The accord reached by Stilwell and Mountbatten did not include questions of 
grand strategy, for differences there went beyond the competence of the local 
commanders. 30 

Immediately after the interview, Mountbatten was seriously injured when 
a bamboo splinter was rammed into his eye as he was driving in his jeep. He 
was temporarily blinded, and had to spend many days in the hospital. The 

27 (1) Stilwell Diary, Notebook 101/2 ■ (2) Stilwell's Mission to China , |p" 379-1 

28 CM-IN 4651, Stilwell to Marshall, 7 Mar 44. 

29 Minute SC4/439/D, SAC for Dill, 16 Mar 44. SEAC War Diary. 

30 Ltr, Mountbatten to Lt Gen Sir Henry Pownall, CofS, SEAC, 8 Mar 44. SEAC War Diary. 



accident was doubly unfortunate, for it immobilized SEAC's commander at a 
time when the Japanese were expected to make their attack on Imphal. 31 

The Chiefs of Staff Reject CULVERIN 

The mission (Axiom) sent to Washington and London in February 1944 to 
persuade Mountbatten's superiors of Culverin's advantages, included Generals 
Wedemeyer and Wheeler of SEAC's American component, and Maj. Gen. 
M. W. M. MacLeod, Capt. M. G. Goodenough, RN, and Brigadier Geoffrey K. 
Bourne. In London the Axiom Mission found that the British Chiefs of Staff 
and the War Cabinet had not settled on what British strategy in the Pacific war 
should be. The military wanted to dispatch a powerful task force to co-operate 
with the U.S. Navy in the Pacific. Mr. Churchill, for the Cabinet, wanted 
British effort to concentrate on the Indian Ocean. The strategic concepts of 
Culverin were most appealing to the Prime Minister but he complained of 
SEAC's estimate of the resources needed. Churchill thought the estimates 
inflated, a "terrific bill." The British Chiefs of Staff objected to the strategy 
expressed in Culverin. They considered an attempt to mount it would post- 
pone the defeat of Japan by six months. As a result of the London discussions, 
the Axiom Mission was told that in Washington they could speak for British 
authority only to the extent that the Prime Minister and Chiefs of Staff agreed 
with Mountbatten's views on Burma operations. Beyond that, no decision had 
been reached. They were also to assure their American hosts that no British 
resources would be withheld from current operations in Burma to mount 
Culverin. 32 

The mission's reception in Washington, March 1944, revealed the Joint 
Chiefs' opposition to Culverin. General Marshall was so impressed with 
Stilwell's objections to Culverin that, instead of just noting their outstanding 
points for the President's consideration, he sent in the whole paper. 33 

The Joint Chiefs objected to Culverin because they believed that resources 
for it would not be available until after the defeat of Germany. Therefore, 
exploitation of a successful Culverin would not reach Singapore before the 
middle of 1945 at the best. This date would be eight to ten months after U.S. 
forces were expected to have cut the Japanese line of communications to that 
area. The Joint Chiefs did not believe that the establishment of Myitkyina as 
an air base would necessitate the recapture of all Burma. While General Wede- 
meyer, Axiom's chief spokesman, agreed that taking Myitkyina would increase 
the Hump lift (in that it would make possible a lower, wider air route to 
China), he added that the operation was too risky. As to north Burma opera- 

31 (1) Mountbatten Report, Pt. B, par. 97. (2) SACs Personal Diary, 7 Mar 44. SEAC War 

32 Rpt of Axiom Mission, 3 Apr 44. SEAC War Diary. 

33 Memo, Marshall for Leahy, 20 Feb 44. Item 251, Bk 15, OPD Exec 9. 



tions in general, Wedemeyer stated: "Without full Chinese help it was impos- 
sible to deny Upper Burma to the enemy, though one of the subordinate 
commanders in the area did consider that it would be possible to reach the 
Myitkyina area. The Supreme Commander and his staff, however, believed this 
to be very problematical, and even if accomplished, would not result in the 
necessary degree of control of Upper Burma. . . ." 

At the end of the SEAC mission's visit to Washington, the Joint Chiefs 
presented a memorandum to the Combined Chiefs of Staff stating they were 
now more firm than ever in their belief that the greatest contribution SEAC 
could make would be to provide timely support to a cross-Pacific advance to 
the China-Formosa-Luzon triangle, and that Mountbatten's greatest accom- 
plishment would be to secure Myitkyina and build up an immediate increase in 
Hump traffic to China. They believed that successes in the Arakan and 
Hukawng areas, plus improvement in the Assam line of communications, 
indicated that difficulties . previously emphasized in Burma operations might 
have been considerably overestimated. 34 

A major change in Japanese naval dispositions supplied the final argument 
against Culverin. In February 1944, the Japanese moved the bulk of their fleet 
from Truk to Singapore. From Singapore it could quickly counter any attempted 
amphibious attack on Sumatra, and the Japanese now had the strength at 
Singapore to make their countermove a success. 3 ^ 

The Japanese Attack Forces Hump Diversions 

The Japanese attack on Imphal had been anticipated, and Slim's plans had 
been long ready. His intention was that 4 Corps should retire before the 
Japanese attack, persuading the Japanese that 4 Corps was in retreat. The 
Japanese would follow over the wild country between the Chindwin River and 
the Imphal plain. There, at Imphal, 4 Corps would turn on the Japanese. 
Heavy reinforcements would be flown in from the Arakan directly to Imphal, 
while a fresh British corps, concentrating around Dimapur, would fall with 
smashing weight on the Japanese flanks and rear. The plan assumed but one 
Japanese regiment would attack Kohima, a town near the Bengal and Assam 
Railway. 36 1 (Map 10) 

54 (1) Mountbatten's understanding of why the JCS wanted Myitkyina taken is given as: 
"They had come to the conclusion that the air effort from China could best be supported by estab- 
lishing air bases and a large staging area at Myitkyina, and by pushing ahead with opening the land 
route into China as fast as possible." Mountbatten Report, Pt. B, par. 135. (2) Wedemeyer's quo- 
tation from Supplementary Min, JCS 154th Mtg, 21 Mar 44, Item 1. (3) CCS 148th Mtg, 3 Mar 
44, Item 1. (4) JCS 774, 16 Mar 44, sub: Strategy in SEAC. (5) CCS 452/10, 21 Mar 44, sub: 
Strategy in SEAC. 

" (1) Mountbatten Report, Pt. B, par. 277. (2) Churchill, Closing the Ring, p. 575. 
36 (1) Ltrs, Col Clarence W. Bennett, Ln O, 4 Corps, to CG, 5 303d Area Comd, 20 Mar, 4 
Apr 44. NCAC Files, KCRC. (2) Mountbatten Report, Pt. B, par. 95. 



The Japanese offensive began on 8 March, a week earlier than had been 
expected, and moved with stunning speed and weight. The Japanese used three 
of their own divisions plus some units (later formed as a division) organized 
among Indian prisoners of war. Four days before the Japanese began their 
attack, Mountbatten had ordered the Manipur front reinforced. 

On 14 March, Mountbatten was told that the 5th Indian Division had not 
yet begun its move to reinforce 4 Corps at Imphal. He also concluded that the 
Japanese attack, though it had been expected, was proceeding with a vigor and 
a momentum thoroughly disturbing to the Fourteenth Army staff. He and his 
staff" had not been aware that the situation was thought to be deteriorating so 
rapidly. Mountbatten believed that responsibility for the nonarrival of the 5th 
Indian Division and for the failure to keep him informed of the situation lay 
with General Giffard, and so informed him. In criticizing Giffard, Mountbatten 
(perhaps because of his recent accident in north Burma) may have been 
unaware that the 5th Indian Division had been launched on an attack on the 
Arakan on 6 March and could not have been disengaged earlier than in fact it 
was. The incident had its effect on the relationship between the two 

Mountbatten personally intervened to speed the reinforcement process, and 
the 161st Indian Brigade was rushed forward. It arrived just in time to find the 
whole of the Japanese 31st Division moving on Kohima. This Japanese action 
was a serious affair, for near at hand, though a stretch of very difficult country 
lay between, was the Bengal and Assam Railway, the line of communications 
to the Hump airfields and Stil well's campaign. 37 

This in turn created another complication for Mountbatten and for Stilwell, 
because in order to fly reinforcements up to save Kohima Mountbatten had to 
divert transports from the Hump, as earlier he had been obliged to do during 
the peak of the Arakan fighting. This diversion of transports, which were the 
sole line of communications of the Fourteenth Air Force, was a serious matter 
for CBI Theater headquarters. Patently, if enough transports were diverted, 
Chennault's position would be endangered, and there would be political reper- 
cussions all the way from Chungking to Washington. Stilwell's deputy, Gen- 
eral Sultan, asked Mountbatten how three Japanese divisions could menace 
Imphal when three Indian divisions had thought it impossible to advance from 
that area. The answer was that the Japanese would take chances such as had led 
to their recent setback in the Arakan, that they normally did not need much 
supply and were entering an area where they could live off the country. Relay- 
ing this explanation to Stilwell, Sultan warned that Mountbatten would divert 
ATC aircraft and place the burden of overruling him on the Combined Chiefs. 38 

37 (1) Rad SEACOS 112, Dial 46, Mountbatten to COS and JSM, 15 Mar 44; Minute SC/ 
486/G, SAC for Giffard, 16 Mar 44; Minute SC4/440/W, SAC for Wedemeyer, 16 Mar 44. SEAC 
War Diary. (2) Ltr, Mountbatten to Dill, 26 Jun 44. Item 70, Folder 57, OPD Exec 10. (3) 
British Comments, Serial 67. 

38 Rad RE 464, Sultan to Stilwell, 15 Mar 44. Item 70, Bk 6A, JWS Personal File. 



Mountbatten did order the equivalent of thirty C-47's off the Hump, and 
the Joint Chiefs at the urgent request of the British agreed to a temporary 
diversion to fly in reinforcements. 39 These transports were a palliative, and on 
25 March Mountbatten wanted to take seventy more C-47's from the Hump. 
The original diversion was being used to fly the 5 th Indian Division from the 
Arakan to Imphal, to be followed by the 7th Indian Division. The request for 
seventy more, made as it was five days before the Japanese cut the Imphal- 
Kohima Road, suggested acceptance of Japanese ability to surround the 5th, 
17th, 20th, and 23d Indian Divisions on the Imphal plain with two divisions of 
their own and to maintain themselves in that position long enough to exhaust 
4 Corps' accumulation of stores and force 4 Corps to go on air supply. Generals 
Sultan and Stratemeyer did not face the prospect of further diversions from the 
Hump with equanimity, and the evident prospect that 4 Corps with 70,000 
combat troops would be cut off by a force that was (mistakenly) believed to be 
much smaller brought increasingly severe but perhaps unjustified criticism 
of 4 Corps conduct of operations. Actually, the Japanese exhibited their 
customary willingness to take chances with logistics by building up the two 
divisions that fought at Imphal to 53,000— 40,000 Japanese, the rest Burmese 
laborers who freed the Japanese to fight. 40 The outcome was that an Army Air 
Forces troop carrier group of sixty-four C-47's was supplied on one month's 
loan from the Mediterranean area as well as an RAF transport squadron of 
fifteen aircraft more. 41 This ultimately permitted return of the twenty C-46 
aircraft diverted from the Hump, when Stratemeyer found they were not being 
fully utilized. 42 

Inkangahtawng: An Attempt That Failed 

Meanwhile, far to the north of Imphal, Stilwell in the Hukawng Valley was 
resolved to try another envelopment of the 18th Division. He was willing to 
accept less than a complete encirclement, for he seems to have contemplated 
putting such pressure on Tanaka's line of communications that the Japanese 
would be obliged to fall back. Merrill and his staff were ordered to make a plan 
to put a roadblock behind the 18th. The orders to Merrill from Chih Hui Pu 

39 (1) Operational Dir 8, Mountbatten to Peirse and Stilwell, 16 Mar 44. Item 285, Bk 3, JWS 
Personal File. (2) Min, SAC's 73d Mtg, 17 Mar 44. SEAC War Diary. 

40 (1) CM-IN 18058, Sultan to Arnold, 25 Mar 44. (2) COS (W) 1237, 26 Mar 44. SEAC 
War Diary, 26, 27 Mar 44. (3) CM-OUT 14633, Arnold to Spaatz for Portal, 26 Mar 44. (4) 
Folder, Wedemeyer-Mountbatten, 3 Sep 43-14 Apr 44, OPD Exec 10. According to U.S. calcula- 
tion, the British disposed of 219 transports in India and the Middle East; these appeared a better 
source than the Hump with all the consequences implied as to Fourteenth Air Force operations. 
(5) Estimated division strengths with attachments: 15th Division— 2 3,400; 31st Division— 34,000; 
33d Division-30,000; service troops-52,000. Of these, the 15th and 33d were at Imphal, the 31st 
at Kohima. Japanese Officers' Comments, 1951 and 1955. 

41 Mountbatten Report, Pt. B, par. 140. 

42 Rad ACC 876, Peirse to Mountbatten, 19 Apr 44; Rad SEAC 284, Mountbatten to Peirse 
and Stratemeyer. SEAC War Diary. 



prescribed a shallow envelopment, and also provided that as at Walawbum the 
Galahad force would be supported by a Chinese regiment that would not be 
under Merrill's command. 43 

Merrill and Sun, personal friends as well as colleagues, discussed these plans 
and concluded they had the same weakness which in their opinion had caused 
difficulties at Walawbum, that is, that the Chinese and American units which 
were to work in the same area were under separate command. Some years later, 
Merrill remembered Sun's offering to let Merrill have operational control of 
any units of the 38th Division that Chih Hui Pu might designate for the 
operation, so as to make effective co-operation possible. Colonel Hunter was 
sent to Stilwell's headquarters to present Merrill's and Sun's recommendations, 
which were to send Merrill's entire force, plus reinforcements from Sun, in a 
wide sweep to set up one block near Shaduzup. Merrill felt that a wider 
envelopment would have less chance of being detected. 44 Meanwhile, the 2 2d 
Division and the tanks would drive down the Kamaing Road from the north. 

Hunter's embassy had only limited success. At the suggestion of his staff, 
Stilwell approved a wider envelopment, but at the cost of splitting the envelop- 
ing force. The orders as finally put out called for the 1st Battalion of the 5307th, 
followed at a day's interval by a regiment of the 38th Division, to make a 
shallow envelopment and block the Kamaing Road south of the pass through 
the Jambu Bum ridge line. The 2d and 3d Battalions, 5307th, followed by 
another Chinese regiment, would make a wide swing to the east, then turn 
back west and cut the Kamaing Road several miles south of the block that 
would result from the shallow envelopment. 45 

In proffering their suggestions, Galahad's staff had been unaware that 
their march could be screened by Kachin irregulars raised by the Office of 
Strategic Services and commanded in the lower Hukawng area by 1st Lt. James 
L. Tilly, and that Lieutenant Tilly had most detailed and accurate information 
of the whereabouts of the Japanese. Because of extremely faulty co-ordination 
and what might be termed a hoarding of information Galahad's staff were 
actually unaware that Tilly's force existed. After the war, Merrill believed such 
data might have permitted Hunter to offer conclusive arguments for the 
Galahad proposals. 46 

Possibly because of concern over the Japanese offensive in the Arakan, the 
Generalissimo attempted to slow Stilwell's advance. 

1. It is believed that you have received my radio to you dated February 23rd. 

2. You are requested to report the operational plans of New 1st Army in advance. 

3. In view of the operations of British forces and the preparations of our forces, the New 
1st Army should not advance until there are advances in Arakan by the British forces. Before 

43 ( 1) Ltr, Merrill to Ward, 26 May 52. OCMH. (2) Scilwell Diary, 21 Mar 44. 

44 Merrill Notes. 

4 ' NCAC History, I, 77. 

46 (1) Merrill Notes. (2) Interv with Tilly, 9 Jun 50. OCMH. (3) Galahad, pp. 49-54. (4) 
FO 11, Hq Chih Hui Pu, 8 Mar 44. NCAC Files, KCRC. 


TROOPS OF MERRILL'S MARAUDERS rest on a mountain trail. 

any advances are made by the friend [sic] forces, our army should stop at the present 
positions, so that we will not be attacked individually. 

4. Please send a reply on the above mentioned two points. 47 

The language was conditional, and Stilwell seemingly resolved to make no 
immediate reply but take the matter up on his next trip to Chungking. 

Stilwell did ask Marshall to have some pressure put on the Generalissimo 
to attack across the Salween into Burma. The Chief of Staff was told that it was 
time the Y-Force attacked, that if just Teng-chung and Lung-ling were taken 
the effort would be worth while. "J ust tnis once can't we get some pressure on 
him?" In view of the performance to date by 4 Corps Stilwell did not think it 
would be safe for him to go below Kamaing. Stilwell hoped that by attacking, 
the Generalissimo might lessen the pressure on Slim and himself. Mountbat- 
ten agreed, and asked his superiors to make such a suggestion to the Chinese 
"with great urgency." 48 

Stilwell also wanted another Chinese division flown to Burma. Mountbatten 

47 Memo, Generalissimo for Stilwell, 6 Mar 44. Item 278, Bk 3, JWS Personal File. 

48 (1) Rad MS 18, Stilwell to Marshall, 16 Mar 44. Item 72, Bk 6A, JWS Personal File. (2) 
Rads, SEACOS 116, Dial 52, 17 Mar 44. SEAC War Diary. 



strongly supported Stilwell's request for it, and also his desire for another 
American long-range penetration group. SEAC's commander asked that the 
President and Prime Minister approach the Generalissimo for another Chinese 
division. 49 

As for the Japanese in north Burma, thanks to the professional skill of Gen- 
eral Tanaka, the foresight of his engineer, Colonel Fukayama, and the fighting 
qualities of their men, the 18 th Division had got away to fight another day. In 
doing so, they had yielded a considerable stretch of north Burma to Stilwell, 
and with every successful withdrawal they backed closer and closer to the Ir- 
rawaddy valley and Myitkyina. Once in the Irrawaddy valley, Stilwell would 
be in a position to advance down the railroad and place himself behind the 
Japanese 13 th Army in its drive on Imphal. Consequently, Tanaka had orders 
to hold the Kamaing area at all costs, which in the Japanese service meant just 
that. So the time was coming when Tanaka would have to stand and fight it 
out. 50 

The terrain in which the operation would take place was the most difficult 
that had yet been encountered. So far the fighting had been in the narrow val- 
ley. Now, in order to make the wide swing, Merrill's force would have to climb 
out of the valley into the hill mass on the eastern side. The hills defy any brief 
description, for the rains have carved a drainage pattern as complex as can be 
imagined. The result is a jumbled collection of hills most difficult to cross in 
any direction. Only one feature offered what might be taken as an avenue of 
approach, the incredibly narrow and steep little valley cut by the Tanai. In the 
area near the Jambu Bum the Tanai flows from south to north, and over the 
centuries has cut a narrow gash in the hills about fifteen miles to the east of 
the Hukawng and Mogaung valleys and roughly parallel. {(Map 

The first day's march of Hunter's two-battalion force went well, though 
frequent rains provided an obstacle and hinted at an early monsoon. A day of 
rest was taken on 18 March, and at dusk an added mission was received from 
Stilwell, to block from the south the approaches to the Tanai valley. General 
Merrill complied the following day by ordering the 2d Battalion plus Khaki 
Combat Team of the 3d Battalion to move south through the hills on the 
Warong trail, reconnoiter the trails toward the important Japanese base of 
Kamaing, and block the Kamaing Road in the general vicinity of Inkangah- 
tawng, while Orange Combat Team of the 3d Battalion stayed near Jan pan in 
the hills and patrolled to the north, south, and west. So sited, Orange Combat 
Team could block the Tanai and still be ready to support the rest of the force 
on the road. The 5307th stayed at Janpan the night of 20 March, received an 
air drop the next day, and held a medical clinic for the Kachins of that vicinity. 
A regular feature of the campaign, these clinics did much to build and hold 

49 Rad SEACOS 118, 17 Mar 44. Folder, Wedemeyer-Mountbatten, OPD Exec 10. 

50 (1) Tanaka Interrog. OCMH. (2) Tanaka Narrative, p. 4. 




the good will of the Kachins, on whom the Americans relied heavily for early 
warning of the Japanese. 51 

On the afternoon of the 21st Hunter and his task force resumed the march 
to Inkangahtawng and the Kamaing Road, with all concerned in the operation, 
from Stilwell down, highly confident. Two days later Hunter was in position 
and decided to put two blocks across the road and patrol to the north. About 
this time, just when he was ready to profit by his march, Hunter learned that 
the 1st Battalion of the 5307th and its Chinese support, which were to make 
the shallow envelopment, had been seriously delayed.- 2 He thought it too late 
to change plans and went ahead. Between 20 and 24 March Hunter, the force 
commander around Inkangahtawng, would have liked to strike directly for 
Kamaing itself with at least three combat teams. His force was at one time 
only about five miles from Kamaing, he had received an air drop, and what he 
learned of Japanese troop movements only confirmed his feeling that Kamaing 
was wide open. But holding it was problematical to Hunter, for with the other 
officers of Galahad he believed that the Chinese division commanders, 
though formally under Stilwell, were in constant communication with the 
Generalissimo, and that the latter had to approve every order of StilwelPs be- 
fore the Chinese would execute it." From this circumstance, in Hunter's opin- 
ion, stemmed what he regarded as a lack of aggressiveness on the part of the 
Chinese, which would make them shy away. However, Hunter placed the at- 
tack on Kamaing before Merrill who replied that Stilwell would not approve 
it. 54 

Merrill's unwillingness to approve such a project seems related to the con- 
clusion he was gradually forming, that with the 22d Division's attack pro- 
gressing so slowly there was a real possibility all three battalions of Galahad 
might be immobilized around Inkangahtawng and take heavy losses while 
accomplishing little that was commensurate. 55 These views he placed before 

Meanwhile, on the scene of action, American patrols encountered the Jap- 
anese on 23 March and reported about a company entrenched at Inkangah- 
tawng. The 2d Battalion went forward as far as possible in face of increasing 
resistance to block the Kamaing Road while Orange Combat Team stayed on 
the east bank of the Mogaung to protect the rear. Shortly after dawn on 24 
March, two reinforced platoons sent to envelop Inkangahtawng found it too 
strongly held and had to withdraw. 

M (1) Galahad, pp. 49-54. (2) Interv with Tilly, 9 Jan 50. OCMH. 
,2 Galahad, p. 54, 55. 

5J Hunter Notes. Hunter actually wrote "Boatner" rather than Stilwell but Boatner had been 
several months in Washington, and as of late March was just returning to his post in north Burma. 
It seems safe to say Stilwell. Later in the campaign, the issue of whether the Generalissimo, 
despite his giving Stilwell command, was interfering in the conduct of operations became a major 
one and it will be discussed below. 

54 Hunter Notes. 

55 (1) Merrill Notes. (2) Stilwell Diary, 24, 25 Mar 44. 



The Japanese reaction to this attempt to block the road, beginning 24 
March, was vigorous, and one Japanese counterattack followed on another. 
These Japanese were a hastily assembled force drawn from a battalion gun 
platoon, an engineer company, a medical company, and division headquarters; 
they were given two 75-mm. guns, placed under command of the 18th Divi- 
sion's senior adjutant, and rushed south to Inkangahtawng. Within the U.S. 
lines Japanese trucks could be heard, presumably with reinforcements; and am- 
munition was running low in the 2d Battalion. At this point, Merrill radioed 
Hunter information coming from a captured Japanese map, that two Japanese 
battalions were moving to outflank him. The message was heard by McGee, 
commanding the 2d Battalion, but not by Hunter. Weighing it, McGee de- 
cided to pull out. He could not tell Hunter because the terrain masked his 
radio. Late on the 24th McGee moved to the Manpin area. 56 

After the 2d Battalion had withdrawn across the river through Khaki Com- 
bat Team, Hunter's men moved to Ngagahtawng and bivouacked there, in the 
area protected by trail blocks and booby traps. 57 While the Galahad combat 
teams were pulling back off the road to less exposed positions, Merrill received 
word from Stilwell "leaving him discretion but telling him to hold back if pos- 
sible." 58 On the afternoon of 26 March Hunter was told that an estimated 
three Japanese battalions were moving north from Kamaing. He called for air 
interception which hit the Japanese forces at 1730, five to eight miles away 
from McGee's force at Manpin. This enemy force was the regimental head- 
quarters and one battalion of the 114th Regiment, organized as two under- 
strength battalions of about 800 men with four infantry guns, under Col. 
Fusayasu Maruyama. 59 His mission was to protect General Tanaka's right flank. 

On the 26th Merrill reported to Stilwell and his staff" that Galahad had 
been withdrawn from the road, Merrill thus having exercised the option Stil- 
well gave him the day before. All morning long Stilwell's headquarters had 
been seeking a way to give timely help but now that Galahad was withdrawn 
Stilwell considered that his presence at the front was no longer required and 
that he was free to visit the Generalissimo in China. He flew there on 27 
March. 60 

Back at Stilwell's headquarters, a captured Japanese sketch suggested that 
a reinforced Japanese battalion was moving north in the Tanai valley to attack 
the flank of the Chinese driving toward Shaduzup. His staff were seriously 
concerned and ordered General Merrill to block this thrust, stopping any move 

56 (1) Memo, Hunter for Greenfield, 11 May 45, sub: Comments on "Merrill's Marauders." 
OCMH. (2)Notes by McGee on draft MS. OCMH. (3) Tanaka, Japanese Comments, Sec. II. (4) 
The Japanese initially reported at Inkangahtawng are described by the Japanese as some soldiers 
passing accidentally. Japanese Officers' Comments, p. 15. 

57 Galahad, pp. 55-59. 

58 (1) Stilwell Diary, 25 Mar 44. (2) Merrill Notes. 

59 (1) Maruyama Interrog. OCMH. (2) Tanaka, Japanese Comments, Sec. II. This left but 
one battalion at Myitkyina. 

60 (1) Stilwell Diary, 26, 27 Mar 44. (2) See|Ch. Villi below. 



north of Nhpum Ga, for there was a good trail through Nhpum Ga and 
Janpan to Shaduzup. This use of the 5307th in a static defensive role was a 
radical change in the concept of its employment. 

With the Japanese moving north from Kamaing, there was the possibility 
that they might thrust themselves between Galahad and Nhpum Ga. Orange 
Combat Team, of the 3d Battalion, under Colonel Beach, was the logical pro- 
tection, as it was nearest the Kamaing Road. Therefore Beach was ordered to 
guard the south flank, which he did with two platoons. 61 McGee, with the 2d 
Battalion and the remainder of the 3d, moved off from Manpin on the after- 
noon of the 26th. McGee set a stiff pace, and kept his men marching until long 
after darkness had fallen. Next dawn they were up and on their way again. 
So quickly and easily described in print, these marches in north Burma were 
in reality exhausting struggles against every variety of obstacle. Staggering up 
stream beds, clawing their way between jungle growths, clinging to hillsides, 
their only respite a few hours on the sodden ground, their food a little K ration 
gulped on the march— the infantry endured a continuous ordeal that language 
is really inadequate to describe. 62 

A trail runs from Manpin to Auche, passing through Warong and Paokum 
in an arc bulging to the south. From Warong and Paokum two trails lead 
south; they would be logical avenues of approach. Another trail from Manpin 
to Auche forms the chord of the arc, and by using it and putting a platoon at 
Warong and Paokum respectively, the two battalions would have their south- 
ern flank well protected as they moved across the Japanese line of advance. At 
noon, 25 March, 1st Lt. Logan E. Weston and his I & R Platoon were ordered 
to Paokum. Lieutenant Weston arrived late that afternoon, set up his road- 
block, and later was joined by a rifle platoon. From Paokum to Warong to 
Auche, Weston, 2d Lt. Warren R. Smith, and their two platoons delayed the 
Japanese on 26, 27, and 28 March, suffering no losses. Covered by Weston's 
and Smith's skillful and well-fought action, the main body arrived at Auche on 
the 27th and left the next morning. 63 

Later, enemy shellfire began and enemy infantry attacked soon after from 
the south. Lieutenant Weston was ordered to fight another delaying action up 
the trail to Nhpum Ga to give the 2d and 3d Battalions time to organize their 
defenses. Shells also fell as the 2d Battalion headquarters and Green Combat 
Team were leaving Auche, the 3d Battalion having gone on ahead. The trail 
ran along a ridge, making it impossible to take cover or disperse, and during a 
brief panic men cast away equipment. One American was killed and perhaps 
five were wounded. 64 The battalion collected itself and, despite the shaking-up 
taken on the trail, quickly set up its perimeter on the hilltop at Nhpum Ga. 

61 Galahad, pp. 58, 59. 

62 Hunter Notes. 

" Ltr, Weston to Col John M. Kemper, Chief, Hist Div SSUSA, 22 Jan 45. OCMH. 
64 (1) Galahad, p. 62. (2) Capt. J. E. T. Hopkins, Medical Study of the 5307th Composite 
Unit, MS. OCMH. 



General Merrill's decision was to split his force. He did not want to dis- 
perse his force strength in that fashion any more than he had wanted to con- 
cur with the original order to put two blocks across the Kamaing Road, but 
in this case he felt the terrain forced dispersion on him. The 2d Battalion was 
put on the hilltop at Nhpum Ga; and the 3d Battalion, three miles north at 
Hsamshingyang to protect the airstrip there. Merrill had one hundred wounded 
who had to be flown out, and Nhpum Ga commanded the airstrip, so it had to 
be held. Both battalions were to patrol to keep the Japanese from bypassing on 
the trail along the Tanai Hka. 65 

The Japanese Delay the 1st Battalion, 5307th 

From bivouac areas in the hills northeast of Shaduzup, on the Kamaing 
Road, the 1st Battalion of the 5307th, followed by the 113th Regiment, moved 
out on the morning of 13 March to put the northern clamp across t he Kama ing 

Road just south of the Jambu Bum, in the vicinity of Shaduzup. (Map 11 and 

12) The I & R Platoon followed some fresh footprints into an enemy bivouac 

and stirred up a hornet's nest of Japanese. There was brisk skirmishing, and 
though the Americans managed to cross the Numpyek Hka just beyond, the 
Japanese had been alerted and proceeded to delay them expertly. Colonel Os- 
borne, commander of the 1st Battalion, decided to cut a fresh trail around the 
Japanese. This was painfully slow business, and waiting for an airdrop took 
another day. On 22 March, when Colonel Hunter to the south was one day's 
march from his goal, aggressive patrolling by Red Combat Team revealed that 
the Japanese had blocked every trail in the area through which Osborne had 
to pass, so again Osborne elected to make his own trail, this time over ground 
so rough that the mules had to be unloaded. The maneuver succeeded, and no 
Japanese were seen on 23 or 24 March. Osborne's march would have been 
greatly aided had he known Tilly's Kachin Rangers were in the same general 

The 1st Battalion's arduous march had placed it close to the Chengun Hka, 
a stream which runs approximately southwest into the Nam Kawng Chaung, 
along whose banks the Kamaing Road runs, south of Shaduzup. Osborne 
elected to follow the Chengun to its confluence with the Nam Kawng and 
place his block across the Kamaing Road, where the chaung makes a U bend 
to the north a few miles south of Shaduzup. 

While the 1st Battalion and the Chinese bivouacked on the banks of the 
Chengun, the I & R Platoon of White Combat Team explored the area where 
Osborne planned to place the block. There they found two Japanese camps 
comfortably placed between the river and the road. There were many Japanese 
at hand, grenading fish, bathing, or loafing under the trees. Bashas filled with 
supplies suggested the establishment was an advance supply depot. Fully in- 

6S Interv with Merrill, 26 Apr 48. OCMH. 



MAP 12 

formed by the I & R Platoon's skillful patrolling, Osborne decided to make 
a surprise attack, sweep the Japanese from the immediate area, and set up his 
block across the road. 

The Galahad battalion and the Chinese 113th Regiment made their way 
down the Chengun Hka on the 27th. Chinese march discipline was not per- 
fect, but on the other hand Japanese security measures were nonexistent. Dur- 
ing the night of 27-28 March the Allied force established itself along the Nam 
Kawng Chaung. At 0300 White Combat Team slipped across the river onto 
the west or Japanese side and crept up within easy reach of the camp. Lying 
on their arms, the men waited for first light and the signal to attack. As the 
Japanese began to build their cooking fires the attack order was given. 

Surprise was complete and overwhelming. With fixed bayonets the combat 



teams swept through the camp. The Japanese scattered in confusion, and the 
area was quickly overrun. Thrusting beyond the camp site, the 1st Battalion 
set up its roadblock, and consolidated its position. The process of consolida- 
tion yielded some dividends of practical value— Japanese rations and equip- 
ment. 66 

The subsequent reaction of the Japanese was much like that at Walawbum. 
They could not risk the blocking of their line of communications while they 
were heavily engaged with the 22d Division some four or five miles to the 
north. The 1st Battalion did not budge, but the Japanese were primarily inter- 
ested in bypassing it to the west and in extricating themselves from between 
Galahad and the 2 2d Division. Later in the day on the 28th, the Americans 
were relieved by the 113th Regiment, which in turn established contact with 
the 22d Division about 30 March. 67 

After its relief the 1st Battalion rested two days, then retraced its route, for 
it was under orders to rejoin the other two battalions which were thought to 
be near Hsamshingyang. Late on 30 March radio orders to proceed to Janpan 
by easy stages were received. The way was rough and the march was very slow. 
Then on 1 April the radio was damaged and Colonel Osborne, worried by 
being out of touch with headquarters, went to Shaduzup and there learned that 
he was to go as quickly as possible to Hsamshingyang. On the 6th word came 
that his help was badly needed. He and his men marched five and one half 
miles that day and night under the worst conditions so far encountered, and 
reached Hsamshingyang at 1700 on Good Friday, 7 April. 68 

Delay at jambu Bum 

While the Galahad battalions and the Chinese 113th Regiment had been 
trying, with varying success, to set up their roadblocks across the Kamaing 
Road and behind the Japanese positions, the 22d Division and the 1st Pro- 
visional Tank Group were attacking the 18th Division from the front. The 
66th Regiment was operating to the west of the Kamaing Road, the 64th Regi- 
ment, to the east, the 65th ( — ) was in division reserve. Directly before the 
2 2d Division was the Jambu Bum, which is about two miles in depth; Shadu- 
zup is five or six miles south of it. To the north of the ridge line is an area of 
paddy fields; to the south the road goes through the thick brush and tall trees 
of the north Burma countryside. A third of the way down the reverse slope is 
the rather considerable obstacle of the Hkawnglaw Hka. 

On 15 March the tanks and the two regiments were about three miles north 
of the ridge line. The 1st and 2d Battalions of the 64th were sent wide round 

66 Galahad, pp. 39-44. 

61 (1) Galahad, pp. 45-47. (2) NCAC History, II, 140. (3) Ltr, Col Van Nana to Ward, 
16 Jul 51. OCMH. (4) Notes by Van Natta on draft MS. (Hereafter, Van Natta Notes.) OCMH. 
68 Galahad, pp. 48-49. 



DAMAGED M3A3 TANKS south of Jambu Bum. 

the left flank. The march was opposed and at one time the lead battalion was 
partly encircled and had to be extricated by a counterattack. Progress of the 
frontal attack was very slow. It was difficult to co-ordinate infantry and tank 
action, because the tankers found it hard to distinguish their countrymen from 
the Japanese. A few such cases of mistaken identity and the 22d's men were 
understandably reluctant to get too close to the tanks. Two days and a sharp 
fight in an open paddy field just north of the crest were needed to reach the 
north slope and crest. 

Once over it, the 22d found the road down to be mined and blocked with 
fallen trees. Two days were lost in clearing the road, and then three battalions 
made a frontal attack, guiding on the road. Next day they tried a co-ordinated 
tank-infantry attack and the leading tank platoon reached the Hkawnglaw Hka 
about four miles south of Jambu Bum, destroying a few machine guns and 
taking four antitank pieces. But the infantry would not follow the tanks and 
dug in two miles short of the stream. The tanks patrolled till dark, then fell 
back to their own lines. 69 

It was the 2/66th's turn to lead on 21 March. Again the tanks got well 
ahead of the infantry. A combination of ambush and counterattack by the Jap- 
s'' G-3 Per Rpts, Hq Chih Hui Pu, 22, 28 Mar 44. NCAC Files, KCRC. 



anese cost five tanks, and then the Japanese came on up the road, almost over- 
running the battalion headquarters, which was saved by the courage of two 
engineer platoons that had been clearing the road. 70 That night the l/64th, 
which had been making the enveloping move, came in from the east and cut 
the Kamaing Road, right in the segment held by the Japanese. A tank attack 
on the 23d found Japanese antitank guns just south of a small stream (north 
of the Hkawnglaw Hka) that prevented the tanks from outflanking the posi- 
tion. The guns knocked out the three lead tanks in quick succession, blocking 
the road. After heavy fighting at the stream crossing, the tanks finally had to 
withdraw, leaving the derelicts. In the afternoon the two flanking battalions 
made their way up the road and joined the 66th. 

Because he did not know of all these delays, the operation seemed to Stil- 
well to be going very well; he believed the 1/5 307th was at Shaduzup, while 
he knew the 2d Battalion was in place at Inkangahtawng. Unfortunately, the 
meeting of the 64th and 66th Regiments did not signal the end of Japanese 
resistance. Japanese and Chinese positions on the road were thoroughly inter- 
mingled, making movement in the immediate area extremely hazardous. An 
attempt to break the deadlock with the tanks failed when the device chosen to 
identify the Chinese infantry backfired. Both the Chinese and the Japanese 
waved white cloths at the tanks. The armor moved blithely on into a nest of 
Japanese antitank men, who destroyed five tanks with magnetic mines, effec- 
tively blocking the road. General Liao, the 2 2d Division's commander, now cut 
a bypass road for the tanks around his west (right) flank. A tank platoon tried 
it, could not cross a ravine improperly prepared for tank crossing, came under 
artillery fire, and had to be withdrawn. 

The decision was now reached to bring the 65th Regiment up from reserve. 
With the l/66th and the 3/64 th attached, it leapfrogged the 64th and 66th 
Regiments on 26 March and attacked. The Japanese held stoutly, counterat- 
tacking five times on 28 March. With the cutting of the Kamaing Road below 
Shaduzup by the 1/5 307th and the Chinese 113th Regiment, Japanese resist- 
ance softened, and on the 29th the 2/65 th entered Shaduzup. In retrospect the 
Chih Hui Pu staff considered, and their judgment seems correct, that their plan 
to establish and exploit the two roadblocks had been thrown out of joint by 
two of the Japanese reactions, the stubborn defense north of Shaduzup (where- 
as it had been thought that the Japanese would conserve their strength), and 
the flanking move up the Tanai valley. There was a third factor, not mentioned 
by Chih Hui Pu, but a sore affliction at the time — unseasonable and heavy 
rains. 71 Looking toward the future, Stilwell, who was highly pleased by the 

70 (1) Unidentified overlay. Folder, X-Ray Force, NCAC Files, KCRC. (2) NCAC History, 
Vol. II. 

71 (1) NCAC History, II, 131-40, and Stilwell Sketches H to K in Vols. I, II. (2) History 
of 1st Prov Tank Gp. OCMH. (3) G-3 Per Rpts, Hq Chih Hui Pu, 5 Apr 44. NCAC Files, KCRC. 
(4) Stilwell Diary, Mar 44. (5) Brown Notes. 



work of the OSS's Kachin Rangers, ordered recruiting and training a force of 
10,000 guerrillas. 72 

By 15 April 1944 the cost to the Chinese of the North Burma Campaign 
was: 22d Division, 800 men killed, 2,000 men wounded; 38th Division, 650 
men killed, 1,450 men wounded. 73 

The Siege of Nhpum Ga 

As previously noted, the tangled jumble of hills that forms the eastern 
boundary of the Hukawng and Mogaung valleys is cut from south to north by 
the deep, narrow valley of the Tanai Hka. If the Japanese made use of this 
valley for a counterstroke, it could lead them around Stilwell's left (eastern) 
flank and into his rear areas. The Japanese force from Kamaing, which had 
compelled the 2d and 3d Battalions of Galahad to yield their roadblock at 
Inkangahtawng, was in a position to launch such a move. Consequently, when 
a captured Japanese sketch had indicated that a reinforced Japanese battalion 
was moving north in the Tanai valley to attack the flank of the Chinese forces 
in the Hukawng Valley, Chih Hui Pu had been seriously concerned, and had 
ordered the 2d and 3d Battalions to stop any Japanese movement north of 
Nhpum Ga. The resulting march had been made just in time, apparently, for 
the Japanese had followed close behind the 2d and 3d Battalions as they moved 
to Nhpum Ga. A well-fought delaying action had given the 2d Battalion time 
to set up a perimeter defense at Nhpum Ga, and the 3d Battalion to establish 
itself at Hsamshingyang, three miles to the north, where it protected an air- 
strip for evacuation and resupply. 74 Their defenses established, the men of the 
2d Battalion had just time to look about their positions before the Japanese 
were on them. 

Nhpum Ga is on a sharp, thin ridge at the northern end of a hill mass. The 
elevation continues for half a mile past Nhpum Ga, then slopes downhill over 
bamboo-covered ground to Hsamshingyang, a clearing in the jungle. To the 
east, the ground drops rapidly to the Tanai, falling 1,400 feet in two miles. To 
the west is the similar valley of the Hkuma Hka, which flows to the south. At 
the northeast quarter of the perimeter established by the 2d Battalion a rocky 
point dominates a water hole, the only local source. The initial perimeter in- 
cluded the water hole in an eastward- jutting salient. 

Both at Walawbum and at Inkangahtawng, Galahad battalions had been 
withdrawn from the roadblocks they occupied. However pertinent the reason 
for these moves, they had not increased the prestige of American arms in the 
eyes of the Chinese, who believed American troops were leaving dangerous 

72 NCAC History, App. 8, p. 1. 

73 Stilwell Diarv. 6 May 44. 

74 See pp. 1181-831 above. 



positions that the Chinese had to take over and maintain. 75 Now, at Nhpum 
Ga, the Americans had orders to stand and fight it out. Tired by their con- 
stant marches, racked by dysentery and malaria, and beginning to show the 
effects of malnutrition, the Americans were now called on to show that they 
too could hold a given patch of ground. 

The Japanese besiegers included Maruyama's original task force and a con- 
siderable body of reinforcements, the main force of the 1st Battalion, 33th In- 
fantry Regiment, which had been the 18th 's right flank guard. Maruyama's 
mission was not to attack the immensely superior Chinese force near Shaduzup 
but rather to push the Americans back as far as possible along the Kumon 
Range in order to protect General Tanaka's right flank. 76 In the first phase, 
28-31 March, the Japanese drove the American outposts on the Auche trail 
back into the Nhpum Ga perimeter on the 28th, then after a brief preparation 
attacked the American position from the south. The attempt failed. Harassing 
fire that night failed to keep the exhausted men of the 5307th awake. Another 
attack from the south came at 0600 on the 29th after an artillery and mortar 
concentration had failed, as did two later efforts in the same direction. The 3d 
Battalion at Hsamshingyang felt unable to help because the need to defend the 
airstrip was vital, but did send a platoon twice daily to keep the trail to Nhpum 
Ga open. On the 30th the Japanese were attacking, ominously from east and 
north. The trail from Nhpum Ga to Hsamshingyang was still open when the 
fighting died down, but next day, attacking from south, east, and northwest, 
the Japanese succeeded in prying the water hole from the Americans' grasp and 
in beating back an attempted counterattack. Moreover, the cumulative effect of 
losses made it impossible to hold so large a perimeter even if the hole had been 
retaken. That same day the patrol from Hsamshingyang found the trail 
blocked, and neither the patrol nor a sortie from the hilltop could restore com- 
munications. Fortunately, the trapped battalion could go on air supply. 77 

Colonel Hunter now commanded at Hsamshingyang. General Merrill had 
long had heart trouble and now was so weak that Stilwell, learning of his con- 
dition, ordered him evacuated over his protests. Merrill was flown out on 31 
March. He did not lay down his responsibilities. On arriving at Ledo, he 
ordered two 75-mm. howitzers sent to Hunter's 3d Battalion by airdrop, be- 
cause Hunter had promptly begun to clear the trail and the forces at his dis- 
posal, Orange and Khaki Combat Teams of the 3d Battalion, needed artillery 
support. Attempts by Orange Combat Team on 1 April failed. On 2 April, at- 
tacking with artillery support and some men from Khaki Combat Team to 
make an envelopment, Hunter again made no progress. Then an attack was 

75 ( 1) For a sample of the Chinese reaction, see Ho, The Big Circle, p. 99. (2) Boatner, who 
at Chih Hui Pu could receive both Chinese and American points of view and had to deal with 
both nations, believed that these withdrawals from the roadblocks embarrassed Chih Hui Pu in 
dealing with the Chinese. See Boatner Notes. 

76 (1) Interrogs, Tanaka, Maruyama. OCMH. (2) Tanaka, Japanese Comments, Sec. II. 

77 (1) Galahad, pp. 62-70. (2) Interrogs, Tanaka, Maruyama. OCMH. (3) Tanaka Narrative. 



made, with air support, by both combat teams and the garrison, but the Jap- 
anese did not budge. 

The situation seemed serious. The Japanese were believed to be moving 
ever more troops up the Tanai and it was believed that the 1st Battalion, which 
had been ordered to aid, would not arrive for four more days at least. Hunter's 
reaction was to attack on the 4th with everyone but the sick and the mule 
skinners, with his large patrols called in and Kachins used to replace them. A 
fake fight, using carbines (which sounded like the Arisaka rifle), was staged 
to deceive the Japanese, and the air support made three passes at them. The 
first two were genuine, the last a feint which made the Japanese take cover, 
only to come out and find the American infantry on them. Hunter's force 
gained that day and came within 1,000 yards of the besieged. About this same 
time Capt. John B. George and a small party, sent north by Hunter to find the 
Chinese regiment which Hunter understood would support him in this area, 
met the 1/1 12th at Tanaiyang, about eight miles northeast. After an interval, 
presumably used to obtain permission to do so, its commander moved toward 
Hsamshingyang. His first element arrived at the airfield on 4 April and was 
used to guard a trail junction. Despite this increasing pressure the Japanese 
made a very heavy attack on the Nhpum Ga garrison, actually reaching the 
foxholes at one point, and being driven out by two soldiers using hand 
grenades. 78 

On its hilltop the garrison, though suffering no shortage of food or ammu- 
nition (aside from the monotony of diet which was itself a hardship), suffered 
from an acute shortage of water. There were no plaster casts for the wounded, 
and they took their sulfadiazine dry. The pack animals could not be protected 
from the Japanese fire. When dead, their carcasses could not be buried, and 
the stench and the carrion flies added more miseries to the battle. Sergeant 
Matsumoto, who had played an important part at Walawbum, was a pillar of 
strength to the garrison, constantly scouting between the lines, overhearing 
Japanese conversations, and informing Colonel McGee accordingly. 79 On one 
occasion, when Matsumoto learned of plans to surprise a small salient at dawn, 
the Americans drew back their lines, booby-trapping the abandoned foxholes. 
Punctually the Japanese attacked, straight into the massed fire of the waiting 
Americans. Throwing themselves into the foxholes for cover, they set off the 
booby traps. Matsumoto completed the debacle by screaming "Charge!" in 

78 ( 1 ) Galahad, pp. 73-80. During the last two weeks of March and the first week of April 
the 114th Regiment remained in reserve in the Maingkwan-Walawbum area. The 113th was 
supporting the l/5307th. The 2d and 3d Battalions of the 112th spent some time in trying to 
find a large group of Japanese stragglers in the Lagang Ga area, failed, returned to Lagang Ga, 
and at the end of the period began to move to the Auche-Warong area. Only the 1/1 12th, re- 
lieved by a battalion of the 30th Division and sent from Pabum down the Tanai valley, took part 
in the battles around the hilltop. (2) Per Rpts, Hq Chih Hui Pu, 28 Mar, 5, 11 Apr 44. NCAC 
Files, KCRC. 

79 ( 1) Stone MS, p. 149- (2) Matsumoto received the Legion of Merit for his services in north 



Japanese, causing a supporting platoon to throw itself on the American guns. 80 
The relieving force made no gains on the 5 th, but the Japanese were begin- 
ning to have their own problems. That day there were only two Japanese at- 
tacks, at 0200 and 0430, which, thanks to Matsumoto, were anticipated and 
stopped. Colonel Maruyama's force was finding its supply problems increas- 
ingly difficult, and Tanaka grew daily more concerned about the defenses of 
Myitkyina. 81 On the 6th Orange Combat Team gained another 200 yards. Two 
key machine guns were knocked out by 2d Lt. William E. Woomer, who con- 
ducted mortar fire from twenty- five yards away. That was the story of the next 
two days, grinding slow advances that gradually cut the gap between the bat- 
talions, and steadily weaker Japanese attacks. 

The long-awaited 1st Battalion of Galahad arrived at 1700 on 7 April, 
after an extended and exhausting march. Two hundred and fifty of its men were 
still capable of exertion. On the 8th and on Easter Sunday they tried to seize 
positions on the trails over which the Japanese moved supplies to their 

On the afternoon of Easter Sunday, the Japanese vanished, leaving cook- 
ing fires and equipment. There was no pursuit, for Stilwell wanted no move- 
ment beyond Nhpum Ga. The battalion of the 114th made its way to Myit- 
kyina, where General Tanaka, anxious about the town, added it to the garrison. 
The attempted roadblock at Inkangahtawng, and the siege of Nhpum Ga, 
which followed when the Americans abandoned the block on 24 March, had 
cost the 5307th 59 dead and 314 wounded. A total of 379 were evacuated by 
air for wounds or illness. The most serious result of Nhpum Ga was the ex- 
haustion of troops. 82 The fighting edge of the most mobile and most obedient 
force that Stilwell had was worn dull. From this fact were to flow consequences 
of great magnitude. There was no compensating damage to the 114th, and it 
bore the major burden of the defense of Myitkyina. 

Air Supply Problem at Imphal 

The consequences of Nhpum Ga were hidden by the veil of the future; at 
the time, the engagement was seen from the point of view of a commander 
whose theater stretched from the Indian Ocean to the Yellow Sea. After check- 
ing with Col. Henry L. Kinnison of Galahad, Stilwell wrote in his diary: 
"Galahad is OK. Hard fight at Nhpum. Cleaned out Japs and hooked up. 
No worry there." 83 Though one of the hardest-fought American engagements 
in Burma, it was from the perspective of CBI Theater headquarters a battle be- 

80 (1) Galahad, p. 78. (2) Notes, Maj John Jones, sub: Burma— Notes on Merrill Expedi- 
tion, 1944, JICA SN 8623, JICA CBI SEA, New Delhi, 27 Sep 44. OCMH. 

81 ( 1) Maruyama and Tanaka comment on the 114th's ammunition supply problem. (2) Ltr, 
Merrill to Ward, 26 May 52. OCMH. 

82 (1) Galahad, pp. 80-84. (2) Merrill's Marauders, p. 94. 

83 Stillwell Diary, 1 1 Apr 44. 



tween a few battalions. Farther south, around Imphal, whole divisions were 
grappling for a prize that might change the course of the war in Asia, and the 
trend of reports from that front was far from reassuring. 

As one report after another on the Japanese advance to Imphal was laid 
before Stilwell, he grew progressively more concerned. Allied defeat at Imphal 
would sweep away everything that he had won. Japanese success in the center 
would redeem every defeat suffered in the Arakan and in north Burma. Esti- 
mating his situation and weighing his resources, Stilwell concluded that he 
could offer SEAC use of the 38th Division to guard his and its line of commu- 
nications and to meet a Japanese thrust across the thirty miles that separated 
Kohima from the Bengal and Assam Railway. At this time he had two of the 
38th's regiments in reserve. To co-ordinate the defense of his line of communi- 
cations he asked Admiral Mountbatten and General Slim to meet with him at 

Jorhat on 3 April. s4 \(See Map 10.) 

At this conference Stilwell was agreeably surprised to learn that General 
Slim, though worried, felt himself master of the situation around Kohima. 
Slim was deploying a new corps, the 33, for defense of that area. Granting that 
the next five to ten days would be critical, Slim wanted no help at Dimapur 
and favored Stilwell's going on to Mogaung and Myitkyina. So did Mount- 
batten, and with great relief Stilwell went back to his campaign. He did accept 
responsibility for local defense of the line of communications from Tinsukia to 
Nazira, with point defense at Jorhat. SOS personnel went through hasty train- 
ing courses, while a striking force was organized. This was Gasper Force— the 
89th Regiment of the Chinese 30th Division, the 2d Battalion, 1st Provisional 
Tank Group, and a heavy mortar battalion. 85 

Soon after the Jorhat conference the Imphal operations made Mountbatten 
fear that expiration of the thirty-day loan period for the Mediterranean aircraft 
would find 4 Corps still dependent on air supply. Therefore, on 22 April he 
radioed General Giffard asking about a request from Generals Slim and Strate- 
meyer to keep the transports until 1 July. Mountbatten found it "very difficult 
to accept" the conclusion that land communications would not be opened be- 
fore then. In reply, Giffard would not promise that communications to 4 Corps 
would be open even by 1 July, and warned Mountbatten that withdrawal of 
these aircraft on schedule would mean disaster for 4 Corps. 86 Therefore, when 
the promised time for their return arrived, Mountbatten was obliged to tell his 
superiors that SEAC had to keep the transports. The CCS acquiesced to the 
extent of extending the loan date to 15 June. By that time the first of a consid- 

84 ( 1) The Stilwell Papers, pp. 285-87. (2) Rad, Stilwell to AGWAR, 31 Mar 44; Rad TK 
20, Stilwell to Marshall, 2 Apr 44. Items 98, 101, Bk 6A, IWS Personal File. 

85 (1) FO 12, Hq Chih Hui Pu, 5 Apr 44. (2) The Stilwell Papers, p. 287. (3) Secy Plans SAC 
(44) 160, Hq SEAC; Min, Conf at Jorhat, 3 Apr 44. SEAC War Diary. (4) Stilwell Diary, 3 
Apr 44. ^^^^ 

86 (1) Se jp. l68,| above. (2) Rad SAC 1779, Mountbatten to Giffard, 22 Apr 44; Rad FE/3, 
Giffard to Mountbatten, 23 Apr 44; Rad OPD 191, Giffard to Mountbatten, 4 May 44. SEAC War 



erable reinforcement of transport aircraft was expected to be on hand, and this 
would permit return of the borrowed aircraft. 87 

Air supply was not a sovereign remedy for the situation created by Japanese 
encirclement of 4 Corps because the transports could not bring in enough. 
Half-rations were ordered, and as operations continued week after week, "it 
was now [in May] calculated that if the rate of air supply could not be in- 
creased 4 Corps was likely to run out of essential supplies and certain supplies 
of ammunition by the first week in July." 88 The outbreak of the monsoon rains 
was imminent in late May, and if this forced a suspension of the airlift, 4 Corps 
might well be forced to attempt a difficult and hazardous withdrawal. This was 
deduced by the Japanese who clung desperately to their positions on the 
perimeter of Imphal. 89 

Why 4 Corps should find itself so close to a defeat that would have incal- 
culable effects in India and China appeared to require explanation, for Allied 
intelligence agencies were not aware that the Japanese had succeeded in mass- 
ing 155,000 men for their attack on India, and their superiors of course were 
no better informed. Because of this error in intelligence, as early as 19 May 
SEAC was calling on Giffard to tell "why with the large number of divisions 
employed the fighting is characterized by company and platoon actions." 
Giffard replied that his first objective was killing Japanese, and that this had 
to be done by destroying each Japanese in his foxhole. Giffard's second objec- 
tive was to break the Japanese encirclement, and his third, to guard the valu- 
able installations on the Imphal plain. There were so many of these, Giffard 
explained, that by the time the security of each had been provided for, the force 
available for aggressive operations had been somewhat reduced. 90 

The Chief of the Imperial General Staff criticized the lack of offensive 
action northward by 4 Corps to link with 33 Corps coming down from 
Dimapur. 91 Sharing the general underestimation of Japanese strength, SEAC's 
deputy chief of staff, General Wedemeyer, commented: 

With a superiority on the ground 2V2 to 1 on the Arakan and Imphal fronts and with 
complete mastery of the air, we remain on the defensive and the enemy retains the initiative. 
Apparently the 4th and 15th Corps have no concrete plans for aggressive and decisive op- 
erations. The battle in those areas may be described as being fought passively on a day-to- 
day basis, countering enemy blows and hoping to bag many Japs in the process. 92 

Wedemeyer believed that only air supply was preventing an ignominious de- 

87 Rad SEACOS 153, Mountbatten to JSM, and COS, 5 May 44; Min, SAC's 95th Mtg, 8 May 
44; Rad SEACOS 162, Mountbatten to JSM and COS, 17 May 44. SEAC War Diary. 

88 Mountbatten Report, Pt. B., par. 196. 

89 ( 1) Rad ACC 953, Peirse to Mountbatten, 20 May 44; Rad 02393, Slim to Giffard and 
Mountbatten, 25 May 44. SEAC War Diary. (2) SEATIC Bull 242, p. 29. MID Library. 

90 (1) Rad 533 OPS, Giffard to MGGS [Major General, General Staff], 19 May 44; Min, 
SAC's 96th Mtg, 9 May 44. SEAC War Diary. (2) Japanese Officers' Comments, p. 15. 

91 Rad SAC 2834, Mountbatten to Giffard, 27 May 44. SEAC War Diary. 

92 Memo, Wedemeyer for Mountbatten, 5 Jun 44. Case 14, OPD 201 (Wedemeyer, A. C. 
[O]), A47-30. 



feat at Imphal. He contrasted the vigorous efforts of 33 Corps to drive from 
Dimapur with the defensive attitude of 4 Corps. To these comments, the ob- 
servations of the NCAC liaison officer with 4 Corps added chapter and verse. 93 
It must be noted that the terrain around Imphal offered several advantages 
to the Japanese. When on the tactical defensive they could occupy steep hills, 
some forested, some grassy. With the antlike industry of the Oriental soldier 
they quickly dug themselves into the hillsides, and soon formidable earth and 
log bunkers faced the Imphal garrison. The Japanese, though on the strategical 
offensive, lay across vital lines of communications; on many occasions they 
could force the defenders to attack Japanese bunkers, while their own attacks 
could be massed on objectives chosen from among a variety of important 

Fortunately for SEAC, Stilwell, and Chennault, the 31st Division never went 
past Kohima into the Brahmaputra valley. Having judged the Japanese by 
European standards, Fourteenth Army had expected them to send only one 
regiment against Kohima. The Japanese had thrown a full division against the 
town with its small and hastily improvised garrison. Thanks to Mountbatten's 
initiative, the whole of the 161 st Brigade was in the area by 29 March. After 
some initial changes in plan caused by an erroneous intelligence report, the 
leading battalion of the l6lst, the 4th Battalion Queen's Own Royal West 
Kent Regiment, Lt. Col. H. J. Laverty commanding, succeeded in entering 
Kohima to reinforce the garrison before the Japanese surrounded the town. 
From the 4th of April until relieved on the 20th, the small garrison of Kohima 
held off the Japanese 3 1st Division. It was a notable feat of arms, and the role 
of the West Kents in the defense of Kohima reflected the greatest credit on the 
regiment and Colonel Laverty. The time won by the gallant defense of Kohima 
permitted 33 Corps to concentrate around Dimapur, to relieve Kohima, and 
then advance to the relief of Imphal. 94 

Relief of Kohima left the Japanese still barring the road from Imphal north 
to Kohima. They held desperately until 22 June when the road was reopened. 
In that time the Japanese made a number of strong local attacks on Imphal, 
which 4 Corps repelled. An offensive effort by 4 Corps was concentrated on 
driving north to meet 33 Corps: . . the opening of this road now became 
our main preoccupation." 95 With the opening of the line of communications 
from Dimapur to Imphal, the enemy effort failed; the Japanese narrowly 
missed a great triumph, but such was the nature of their gamble that their 
failure resulted in disaster for them. 

93 (1) Ltrs, Col Bennett to CG, 5303d Area Comd, sub: Daily Int and Opns Sitreps. NCAC 
Files, KCRC. (2) Japanese Officers' Comments, p. 15. 

94 ( 1) Maj. E. B. Stanley Clarke and Maj. A. T. Tillott, From Kent to Kohima, Being the History 
of The 4th Battalion the Queen's Own Royal West Kent Regiment (T.A.), 1939-1947 (Aldershot, 
England, 1951), Ch. XI. (2) Operational Record of Eleventh Army Group and A.L.F.S.E. A., 
November, 1943-August, 1945, pp. 27-28. OCMH. (3) Japanese Comments, Sec. I. (4) Mount- 
batten Report, Pt. B, pars. 93-116. (5) Owen, The Campaign in Burma, Ch. 14. 

95 Mountbatten Report, Pt. B, Par. 192. 



The Japanese, with three divisions plus the Indian nationalist formations, 
now found themselves at the end of muddy jungle trails with the monsoon 
rains pouring down. The Japanese supply system had never brought up food to 
the 31st Division at Kohima or to the 13th north of Imphal. Both relied on 
what they could extract from the hill folk. Supply arrangements for the 33d 
Division began to break down in May. Its food could be brought forward some 
distance up the Japanese line of communications from the Chindwin, but the 
condition of the trails, the rains, and the attentions of Allied aircraft made 
distribution almost impossible. 

In the last week of June, General Mutaguchi concluded that the breaking of 
his blockade of Imphal on 22 June, the extremely bad Japanese supply situa- 
tion, and the exhaustion of the fighting strength of his force left no hope for 
success. His conclusions were presented to his superiors in Burma Area Army 
and Southern Army. Perhaps ten days passed in arranging for the issuance of the 
order to suspend operations. 

The intent of the order that Gen. Masakazu Kawabe, Burma Area Army 
commander, issued about 5 July, was not fully grasped by his subordinates, and 
so 13th Army did not discontinue its attacks until 15 July. The monsoon rains 
were then falling, and the state of the Japanese soldiers was soon tragic. The 
Japanese were exhausted by malnutrition and malaria. Roads collapsed in mud. 
Over 17,000 horses died. Morale broke, and the Japanese, who had done all 
that soldiers could, now began to throw away their arms, quarrel, and fight for 
food. Only the 33d Division kept its spirit and acted as a rear guard in the long 
retreat back to the Chindwin. 

Of the 155,000 Japanese combat and supply troops who engaged in the 
drive on Imphal, 65,000 died. Deaths from sickness and hunger were far more 
numerous than battle casualties, and many of the survivors were diseased. After 
the war, the Japanese set the casualty rate in combat units at 85 to 90 percent. 
The U operation was a debacle and ruined the officers who directed it. Relieved 
were: the Commander in Chief, Burma Area Army, and his chief of staff; the 
commanding general and chief of staff of the 13th Army with their senior staff 
officers; the division commanders and their chiefs of staff. 96 The battles around 
Imphal were decisive for the Japanese in Burma. They could still delay, but 
they could not hope to win. 

Meanwhile, on the British side, the victors of Imphal were resolved to press 
their hard-won advantages. The question facing Mountbatten, Slim, and 
Giffard was that of choosing the better way to exploit success— whether to 

96 (1) Lt. Gen. Shibata Uichi, who assumed command of the 15th Division on 30 June 1944, 
remarked that in the retreat he had to abandon large stocks of food which he could not distribute. 
His forward elements ate what was left of their original stores, supplementing their rations with 
grass, and sometimes attacking 4 Corps positions to get the rations to be found in them. SEATIC 
Bull 240, pp. 7, 10-13, 23-24, 51; SEATIC Bull 242, pp. 29, 31-32; SEATIC Bull 245, p. 7; 
SEATIC Bull 247, pp. 23, 25, 26-28. MID Library. (2) Japanese Comments, Sec. I. (3) Jap- 
anese Officers' Comments, App. 2, Mutaguchi; App. 4, Kawabe. 



undertake a close pursuit of the beaten Japanese to the Chindwin, or to request 
from the CCS the air and amphibious resources for a stroke against some 
strategic point in Japanese-held territory. The choice was not an easy one. Close 
pursuit through the monsoon rains meant sending tired battalions, their ranks 
thinned by wounds, death, and sickness, to force a way along the same sort of 
muddy hillsides and half-drowned valleys that had contributed so heavily to 
the Japanese defeat. On the other hand, waiting for amphibious resources 
might give the Japanese time to repair the defeat. The alternatives were simple, 
but immensely difficult to weigh. 

The Chindits Go Back to Burma 

While the Japanese were hurling themselves on Imphal, and while Stilwell 
was gathering his Chinese to close on Kamaing and Myitkyina, a third active 
front was opened in central Burma. Despite SEAC's belief that too deep an 
involvement in Burma was unwise, and despite the changed situation caused 
by the Japanese offensive, that portion of the old SEAC plans which called for 
General Wingate and his men to be dropped into central Burma on the 
Japanese lines of communications was put into operation on 5 March. 

For his greatest experiment in long-range penetration Wingate had 
assembled and trained a formidable force, largely drawn from the 70th British 
Division, a veteran formation with long Middle East experience. This force was 
given the cover name of 3d Indian Division though it included few Indian 
troops. The division had six brigades, 14th, 16th, 3d West African, 77th, 
111th, and 23d, each of four battalions. Because of the Imphal crisis 23d 
Brigade was later detached, leaving five brigades. The Chindits had their own 
pocket-sized air force, designated as No. 1 Air Commando, AAF, and including 
fighter bombers, medium bombers, transports, and liaison craft. 

The mission of Wingate's force changed not long after it entered Burma. 
The personal intentions of the aggressive Wingate were in line with the 
original SEAC plan chosen by Mountbatten so many months before: to 
establish an airborne division at Indaw before the monsoon rains and link up 
with it later. With the cancellation of Buccaneer and the subsequent 
anticipation of the Japanese drive on Imphal this plan had been forgotten by 
all but Wingate. 

General Wingate had the highest hopes, though he made no promises, that 
by concentrating three brigades around Indaw, he might force the Japanese 
to evacuate north Burma. He hoped to threaten Japanese communications 
south of Wuntho, menacing the 15th Army and giving 4 Corps a chance to 
turn the tables with a speedy advance to the Chindwin. Then "14 Brigade (less 
two columns) can be introduced to Pakokku to sever the 33d Division's com- 
munications, and further exploit against divisions to the north. This should 


ermit a bold advance by 4 Corps raftwise down the Chindwin to create a 
ridgehead in the dry zone prior to the monsoon." If operations went well, 23d 
Brigade would be put into the Metktila area to exploit the situation generally. 
This done, the Japanese at the beginning of the monsoon would face a front 
stretching from Pakokku to Lashio with the Chindwin in British hands. 
Wingate's aims were thus far different from Slim's plan to lure the Japanese to 
Impha! and there defeat them. He saw them recoiling in confusion, stunned 
and baffled by the conflagration at their rear, and then hotly pursued across the 
Chindwin— all before the monsoon began in mid-May. g7 

The fly-in began late on 5 March. The whole of 77rh Brigade was directed 
onto one airstrip, BROADWAY, about fifty miles northeast of Indaw, as the 
Japanese had blocked one of the other sites contemplated, and it was feared that 
time did not permit briefing pilots for the third. Among the first in were U.S. 
engineer personnel who, despite repeated glider crashes and pile-ups m the 
confined space, shook themselves clear and had their strip ready for the Trans- 
ports. In came the rest of the 77th Brigade, soon followed by the 111th. The 
fly-in was complete by 11 March, and rhe 9,250 Chindits proceeded to organize 
themselves with their own airstrips for fighter cover and resupply, and their 
own 25-pounder artillery. Moving over to the railway, 77th Brigade established 
a stronghold near Mawlu, White City, * and cur the line of communications 
supporting the 18th Division far to the north. This action was not, of course, 
immediately decisive, for the 18 th had been accumulating supplies for a year, 
but it meant ultimate defeat for the 18th Division if the British could not be 
driven away." 

The rime chosen to drop the Chindits near his line of communications was 
most embarrassing to General Tanaka, On 10 January 1944. Headquarters, 15th 
Army, had suspended movement of supplies to the 18th Division in order to 
accumulate stocks for the projected attack on Imphal. Shipment was to resume 
as soon as the Imphal operarion was under way. Then the Chindirs cut the rail 
line, and just when the supply movement was to have resumed, the 18th Divi- 
sion had to start living on what was at hand in north Burma. It was aided by 
the 56th Division, which shipped about ninety tons of vital supplies to 
Myitkyina via Bhamo, but Tanaka's supply position was fundamentally 
compromised by the Chindit fighting along the railway to north Burma. 10 " 

The landings created panic among Japanese line of communications troops. 
A scratch force was hastily gathered to deal with the Chindits, for the effort 

97 ( l ) Se dlh. ILfe bove. \ 2 ) App reciation COS/24, Comdr Special Force at Imphal [Wingate] 
tor SACSEA, 10 Feb 44. sub LRPG Opns, Exploitation of Opri Thursday, Memo, Comdr Spe- 
cial Force for SACSEA, 13 Mar 44, sub: Forecast of Possible Development of Thursday. SEAC 
War Diary- 

?■ So called both from the supply parachutes that festooned the trees, and after an amusement 
park of that name. 

{ 1) Mountbatten Report, Pt. B, pars. 1 17-25. {2} Tanaka, Japanese Comments, Sec. lit 
100 Tanaka, Japanese Comments, Sec. Til. 



was underestimated. Then, when the first Japanese attempts to clear the 
Chindits from the railway were unsuccessful (around 16 and 17 March), some 
staff officers of the Japanese 15th Army were apprehensive and suggested ending 
the offensive on Imphal, but General Mutaguchi resolved to continue it and 
instead directed enough troops against the Chindits to deal with them. So odd 
battalions came hurrying from all over, the 2 /31st of the 15th Division, 3/1 14th 
of the 18th Division, 1/ 148th of the 56th Division, and 1 /213th of the 33d Divi- 
sion. Each was thus drawn away from a parent unit at a key sector, for example, 
the 3/ 114th, which had just been released from army reserve to join the 18th 
Division. The 24th Independent Mixed Brigade (less one battalion) was moved 
north and opened headquarters at Indaw on 25 March. Farther back, the 53d 
Division, then moving slowly into Burma, was ordered to march north and 
engage as fast as its units arrived. All this, of course, was a diversion from the 
main effort toward Imphal, and a great aid to Stilwell's operations. The 111th 
Brigade paid its respects to the Imphal operation by harassing the Japanese 
line of communications, Indaw-Tamu, from 10 March to 30 April, forcing the 
Japanese to abandon its use. 101 

By mid-March the Japanese offensive seemed to SEAC to need stronger 
measures, so the fly-in of the rest of 3d Indian Division, originally intended 
to relieve the first wave, was accelerated. The 3d West African and 14th 
Brigades were flown into a new stronghold (Aberdeen) near Manhton, and 
arrangements were made for the 16th Brigade which, against fantastic diffi- 
culties of terrain, had marched in all the way from Ledo, past the fighting in 
the Hukawng-Mogaung valleys, without pause except for a diversionary 
attack on Lonkin. General Wingate now had his five brigades deep in central 
Burma; but before he could seize the rewards of his planning, thought, and 
preparation, he was killed in an aircraft wreck on 24 March. His death was a 
heavy blow. After the war the commander of the 16th Brigade explained the 
unique difficulties in which Wingate's death placed his men: 

General Wingate's death would have been a catastrophe whenever it happened, but it 
could not have happened at a worse moment than it did. He was on his way out from a 
conference with myself and another of his brigade commanders; the orders which he had 
given us were unknown to his headquarters, and their ignorance of them was unknown to 
us. His successor was in the field, and could not be reached; the headquarters themselves 
were on the move from Imphal (which was nearly threatened by the Japs) to India, with 
results to our communications which may be imagined; and a further blow to our radio 
contact took the form of four days of violent thunderstorms. These days of confusion left a 
legacy behind them which made its mark and contributed largely to the breach of two 
principles of war which were all-important to us: those of concentration and maintenance 
of the objective; though in the latter case, breach of principle was inevitable with the change 
of plan. 

101 (1) SEATIC Bull 244, History of Japanese 33d Army, 3 Oct 46, pp. 19-20, 71-72; SEATIC 
Bull 245, pp. 3-4; SEATIC Bull 247, pp. 11-13. MID Library. (2) Cyril N. Barclay, The History 
of the Cameronians (Scottish Rifles) (London, 1947), pp. 147-48. (3) Japanese Comments, Sec. I. 



At the time, and indeed until I discussed this lecture with General Slim, I and all the 
brigade commanders were under the impression that the change of plan happened at this 
stage when the Japanese were developing their movement across the Chindwin. But now 
I find that the plan was actually altered some weeks or months before, when the Japanese 
intentions first became apparent. For reasons of his own this decision [fall back to Imphal] 
was not communicated to us by General Wingate, perhaps in the hope that he might be 
able to win such success as would justify the High Command into reverting to the original 
plan. 102 

The next development in Stilwell's relation to the 3d Indian Division was 
at the Jorhat conference of 3 April. There, it was agreed that as Stilwell con- 
tinued to advance on Mogaung and Myitkyina two brigades would help his 
advance by operating on the Japanese line of communications while two others 
would head west across the Chindwin to harass the Japanese lines of com- 
munications to Imphal, if they could cross the Chindwin's escarpment. Thus, 
instead of thirteen battalions operating vigorously behind 15lh Army and 
slicing its communications, the 3d Indian Division had its strength dispersed 
among several forces with no clear-cut mission. 

Shortly after the Jorhat conference the 3d Indian Division received a new 
directive ordering it to devote its principal efforts to helping Stilwell, once it 
had completed the successful evacuation of its initial strongholds and landing 
strips. This employment of the 3d Indian Division speedily led to differences 
between Stilwell and Maj. Gen. W. D. A. Lentaigne, Wingate's successor. 
Stilwell wanted the LRPG's to hold firmly at Indaw, cutting the Japanese 
line of communications to north Burma and keeping major Japanese forces 
away from his right flank. Against this was SEAC's belief that the endurance 
of the Chindits was limited to about ninety days, which meant taking them 
out in mid-June and moving the division at a still earlier date to a place from 
which it could be flown out. Proposals for their withdrawal from Burma would 
therefore inevitably be made by the Chindit commanders sometime in June, 
and maneuvering to an assembly area might well take them off the railroad. 
Either would be bitterly opposed by Stilwell, who did not, as he told General 
Sultan, relish the prospect of Lentaigne's men retiring on him and bringing in 
their wake all the Japanese that they had attracted. 

A British Broadcasting Corporation announcement that Stilwell and 
Lentaigne were in perfect accord brought the matter before higher authority 
when Stilwell denounced it to Mountbatten and Marshall as a fabrication. He 
would accept the British leaving Mawlu, he said, since higher authority had 
directed it, but he did not like the public proclamation that it was in accord 
with his views. 103 

102 B. E. Fergusson, "Behind tne Enemy's Lines in Burma," Journal of the Royal United Service 
Institution (August, 19 46). 

103 (1) Papers cite J n. »S (i)J (2) Rad SH 40, 16 Apr 44; Rad SH 44, Stilwell to Sultan, 17 
Apr 44; Rad SH 74, Stilwell to Sultan, 24 Apr 44. Items 117, 120, 140, Bk 6A, JWS Personal File. 
(3) Memo, Wingate for War Cabinet, 10 Aug 43. SEAC War Diary. (4) Fourteenth Army Opn 
Instr 60, 4 Apr 44. SNF 4. (5) Mountbatten Report, Pt. B, par. 165. 



The Question of Myitkyina 

While the British, Gurkhas, and West Africans of the 3d Indian Division 
were cutting the Japanese line of communications to Myitkyina, and Stilwell's 
Chinese were advancing on the town, the highest Allied authorities were 
engaged in discussing whether Myitkyina's capture was worth while and should 
be attempted. Following the visit of the SEAC mission to Washington and 
London and the inability of the mission to secure approval of Culverin, the 
question of a directive for SEAC was still unsettled. In late March 1944 the 
Joint Chiefs again urged that the Combined Chiefs direct SEAC to undertake 
vigorous operations in north Burma. Because militarization of the Bengal and 
Assam Railway had greatly increased the capacity of the Assam line of com- 
munications, and because recent operations in the Arakan and the Hukawng 
Valley had gone so well, the Joint Chiefs hoped SEAC would be more favor- 
ably inclined toward seizure of north Burma. Apprised of theJCS view through 
channels, SEAC's planners asserted that nothing had happened to change their 
belief that it was an extremely uneconomical use of manpower and resources to 
become deeply involved in the Burmese jungles. 104 The SEAC planners always 
hoped that the CCS would assign to SEAC air and sea resources that would 
permit the Allies to exploit successes at Imphal by airborne or amphibious 
assault— which the Japanese would have been hard put to counter— against 
vital points behind the Japanese defenses of Burma, such as Rangoon. 

Admiral Mountbatten, basing his reply to the JCS recommendations on his 
planners' views, stated that a preliminary examination of the possibility of 
taking the Mogaung-Myitkyina area and dominating north Burma down to the 
line Katha-Bhamo by December 1944 showed that two more infantry divi- 
sions, a parachute brigade, and a long-range penetration brigade should be on 
hand in India, and two more divisions should arrive by the winter of 1944-45. 
"My conclusion is that the conquest of Northern Burma down to the line 
Katha-Bhamo is impossible to carry out by the given dates, and that even by 
later dates it is unsound and should not be attempted." Rather than suggest 
Culverin again, Mountbatten proposed an operation against the Prome- 
Rangoon area after the main thrust in the Pacific. References to holding a 
perimeter in the Imphal area after the conclusion of current operations there 
suggest that as of mid- April SEAC was not yet committed to exploiting victory 
at Imphal by a major thrust across the Chindwin into central Burma, but 
preferred an amphibious attack on Rangoon. 105 

Through CBI headquarters in Delhi, Marshall repeated to Stilwell the JCS 
view that Myitkyina had to be seized and a buffer zone created to the south so 

104 (1) Rad, JSM to SACSEA, 24 Mar 44; Secy Plans SAC (44) 166/1, Hq SEAC, 10 Apr 44. 
SEAC War Diary. (2) CCS 452/10, 21 Mar 44, sub: Strategy in SEAC. 

105 Rad SEACOS 137, SACSEA to War Cabinet and JSM, 14 Apr 44. ABC 384 (Burma) 8-25- 
42, Sec 6, A48-224. Also in JWS Misc Papers, 1944. 



that Hump tonnage could be increased. The build-up of Hump tonnage to 
20,000 tons a month and completion of a two-way, all-weather road and a 
4-inch pipeline to China seemed necessary and timely to Marshall so that the 
maximum possible aid to projected U.S. operations in the Pacific might be 
given by China-based air power in early 1945. Commenting on these orders 
from Washington, General Stratemeyer pointed out to General Arnold that 
Myitkyina as an oil head at which Hump transports could refuel was the key 
to the sharp increase in Hump tonnage that was the prerequisite of any 
ambitious plans in China Theater. 

Both Stratemeyer and Sultan observed that plans for Burma operations 
ultimately depended on a Combined Chiefs decision on strategy. Sultan stated 
that there was an inclination in SEAC not to accept support of the main Pacific 
thrust as its primary mission if such a mission required land operations in 
north Burma, because SEAC thought such operations not worth the effort. 
Sultan believed SEAC would propose Hump expansion without the commit- 
ment of any of its ground forces to Burma, in order that the campaign to secure 
the Mogaung-Myitkyina area might resolve itself into a Sino- American effort 
entirely. To support this thesis, Sultan pointed out that the current plans for 
the Chindits called for their withdrawal through the Mogaung valley, which 
would leave Stilwell without flank support. 106 

A new directive, foreshadowed by Marshall's recent radio to Stilwell, 
arrived in CBI Theater on 3 May 1944 direct from the Joint Chiefs of Staff. 
It gave CBI Theater a new mission, emphasized the importance of Myitkyina 
in the plans of Stilwell's American superiors, and relegated to the past any 
question of whether CBI Theater headquarters favored carrying out the 
Sextant decisions or of how it reacted to Mountbatten's views. The JCS now 
ordered CBI Theater to clear north Burma and take Myitkyina in order to carry 
out the new mission of CBI— air support to forthcoming operations in the 
western Pacific. Land-based air support from Chinese bases was declared to be 
necessary for successful carrier aircraft operations against Formosa, the Ryukyu 
Islands, the Philippine Islands, and the China coast. Stilwell was also ordered 
to give what indirect support he could to attacks on Mindanao Island of the 
Philippine group in November 1944, without prejudice to operations of greater 
strategic importance, such as taking Myitkyina. The Joint Chiefs asked Stilwell 
for an estimate of the effort he could make to support these Pacific operations 
and to cover temporary reinforcement of the Fourteenth Air Force by the Tenth 
Air Force. The JCS said they were aware that accumulating logistical support 
for the projected air effort would require major curtailment of Hump support 
for the Chinese ground forces in China and of other activities that did not 

106 (1) Rad SH 3, Stilwell to Sultan, 7 Apr 44; Rad CRA 1416, Sultan to Stilwell, 16 Apr 44. 
Items 111, 118, Bk 6A, JWS Personal File. (2) CM-IN 19350, Sultan to Marshall, 27 Apr 44. (3) 
Rad CRA 2367, Sultan to Marshall, 30 Apr 44. SEAC War Diary. 



directly support the air effort. The immediate and progressive stockpiling of 
supplies in China was ordered. 107 

As Sultan had predicted, the SEAC planners on 6 May reported that the 
Mogaung-Myitkyina area could be taken by accepting very grave risks else- 
where. Should these, which they tabulated at length, be unacceptable, then they 
suggested the old plan of a less advanced oil head. But in the light of the JCS 
directive these views could not be controlling on CBI Theater. The plan that 
CBI promptly submitted to the Joint Chiefs and SEAC called for, as Phase I, 
seizure of the Mogaung-Myitkyina area as soon as possible; Phase II, prepara- 
tion of all-weather airfields there by 1 January 1945, and clearance of the trace 
of the road from Myitkyina to Kunming in order to increase support for the 
ground and air forces in China; and Phase III, completion of the India-China 
line of communications before the end of 1945. 108 When these radios were sent, 
StilwelPs men were ten days' march from Myitkyina. 

To capture the Mogaung-Myitkyina area Stilwell would have the use of five 
Chinese divisions. Personnel of the 30th Division had been trained at Ramgarh 
as they arrived at wide intervals from July 1943 on. The 88th Regiment of the 
30th Division left Ledo for Shingbwiyang on 1 March 1944. By 24 March its 
forward concentration was complete and it was on its way to Pabum to relieve 
the 1/1 12th, 38th Division. The 89th Regiment, at Jorhat to guard against a 
Japanese thrust past Kohima, would be ready as soon as 33 Corps was assembled 
to relieve Kohima. 109 

When Galahad's withdrawal from about Inkangahtawng rendered imme- 
diate encirclement of the 18th Division a remote possibility, Stilwell flew to 
Chungking and asked the Generalissimo for reinforcements in order that the 
success already gained might be exploited and Myitkyina seized. On 28 and 
29 March Stilwell met with the Generalissimo. No conference minutes are in 
Stilwell's files, but his diary records that the Generalissimo agreed that the 
Chinese 50th Division might be flown to north Burma. Stilwell asked for 
another division, but the decision was postponed for a few days. Then he was 
offered the Chinese 14th Division. Stilwell's first reply was that only the 50th 
Division could be airlifted to Maingkwan, where an airstrip had been built, and 
he asked that the Chinese leader consider infiltrating the 14th Division through 
the Kaoli-kung Mountains to attack Myitkyina from the China side. Then he 
reconsidered and on 1 April asked General Hearn to have Gen. Ho Ying-chin 
consent to fly both the 50th and 14th Divisions over the Hump. The Chinese 
moved with commendable promptness. By 15 April 1944 almost all of the 50th 
Division had been airlifted to Maingkwan, and that same week the 14th Divi- 

107 Rad CRA 2655, Gen Evans to Stilwell, 4 May 44. Item 102, Bk 6A, JWS Personal File. 
This radio was a relay of the JCS directive to Stilwell at the front. 

108 Revised Rpt, War Stf on WS (44) 29, 6 May 44; Rad CRA 2927, Sultan to Marshall, 7 
May 44. SEAC War Diary. 

109 G-3 Rpts, Chih Hui Pu, 7-28 Mar 44. NCAC Files, KCRC. 



sion began to arrive at the Hump airfields. These were of course transfers of 
men, not of fully equipped units ready to fight. 110 

Thus, by mid-April 1944, of 316 Chinese divisions, five had been committed 
to operations in north Burma. Opposing those five were three understrength 
regiments of the Japanese 18th Division, soon to be reinforced by two regimen- 
tal headquarters and three battalions. The arms, uniforms, food, pay, and 
ammunition supplied to the five Chinese divisions in north Burma were U.S. 
lend-lease and British reciprocal aid. The medical and service units supporting 
them were from the U.S. Army. Air support— tactical, strategic, supply, and 
evacuation— was given by the Army Air Forces. 111 


When, in May 1944, Marshall ordered Stilwell to take Myitkyina, there was 
no longer any possibility that Stilwell's superiors might of their own initiative 
order the North Burma Campaign to be canceled. When Kohima was relieved 
on the 20th of the month the Japanese offensive, though still very menacing, 
was no longer an immediate threat to the line of communications that sup- 
ported operations in north Burma. On 29 March the Chinese had entered 
Shaduzup, which meant they were over the Jambu Bum and had opened the 
door to the Mogaung valley, at whose south end lay the Irrawaddy and the 
Burma railway and, a few miles beyond, Myitkyina itself. The campaign was 
no longer in the balance; both in the council room and on the field, the scales 
had inclined. 

110 (1) Stilwell Diary, 28, 29 Mar 44. (2) G-3 Rpts, Chih Hui Pu, 5-18 Apr 44. NCAC Files, 
KCRC. (3) Rad CFB 15536, Dorn to Stilwell, 31 Mar 44; Rad SHC 1, Stilwell to Hearn, 1 Apr 
44. Items 2152, 2154, Bk 6, JWS Personal File. 

111 For a detailed account of the organization, training, and equipping of the Chinese Army in 
India, see Stilwell's Mission to ChinaTCKapteFVLl 


The Drive for Myitkyina 

While the Chinese Army in India had been edging up to Shaduzup the 
thought had crossed Stilwell's mind that Shaduzup might be as far as his forces 
could get before the rains began. Then the Japanese drive on Imphal began to 
acquire a disturbing aspect, and the conference between Mountbatten, Slim, 
and Stilwell was called at Jorhat on 3 April 1944. At the conference, Slim 
expressed his confidence that he would win at Imphal. Perhaps as a result of 
that confidence, Mountbatten confirmed the existing directives that called for 
Stilwell to take the Mogaung-Myitkyina area. Meanwhile, in one radio after 
another, SEACs staff told the Joint and Combined Chiefs that Myitkyina 
probably could not be taken without sending heavy added reinforcements to 
SEAC, if taken probably could not be held, and even if held was not worth 

The Japanese offensive on India, the slow progress of the North Burma 
Campaign, the Generalissimo's reluctance to cross the Salween, and the steady 
consumption of time, all registered on Stilwell's estimate of what he could do 
in north Burma. His estimate of what he could do with the means his several 
superiors had allotted him began to shrink drastically. 


Time was pressing on Stilwell as he in turn was pressing on General 
Tanaka; if he was to take Myitkyina he must take it with a quick bold stroke 
before the rains began. The solution began to reveal itself to Stilwell immedi- 
ately after the Jorhat conference. Over the next few days the solution became a 
plan and was presented to his Chinese subordinates and to Merrill. 

To seize the Mogaung-Myitkyina area as directed by Mountbatten at Jorhat, 
Stilwell determined to drive down the Mogaung valley on Kamaing with such 
vigor as to persuade General Tanaka that this was the principal effort. Mean- 
while, a task force of Galahad's survivors (End Run Force) plus two Chinese 
regiments and a Kachin screen would slip east over the Kumon Range, which 
formed the eastern boundary of the Mogaung valley, and strike directly at 



Myitkyina. Myitkyina's garrison would, Stilwell hoped, be depleted to help 
defend Mogaung from the Chindits. 1 

In driving for Myitkyina, Stilwell's immediate objective was to increase 
Hump tonnage. As of this moment, he did not believe that opening a land 
line of communications to China, which he thought to be still the mission 
given him by the Trident, Quadrant, and Sextant Conferences, was any 
longer within his capabilities. The Sextant directives and the radios from 
Marshall had stressed the connection between Myitkyina and the Hump. 
Stilwell ordered Hearn to: 

Make following report to the Generalissimo. Without holding Mogaung he will of 
course realize that it will be impossible to build road and pipeline into Myitkyina and as a 
result the anticipated increase in Hump tonnage will not materialize. If we had Myitkyina 
and a pipeline into it, the Hump tonnage could be materially increased. In an effort to get 
forward and make this possible I have issued orders for an advance to certain areas making 
use of maximum force and effort. These orders are not being put into effect with the speed 
and effort which in my opinion the situation requires. As a result, the chances of getting to 
Mogaung are now slim and any increase in Hump tonnage will disappear, if these chances 
dwindle further. I am reporting this so that the Generalissimo will know that I am not 
satisfied with what the division commanders are doing to get forward and make our mission 
a complete success. 2 

A few days later, in commenting on a plan for major airborne operations in 
Burma the next fall, Stilwell observed that the decisions of the Trident, 
Quadrant, and Sextant Conferences still held: . . to get enough of 
No. Burma to reopen communications with China. ... It could have been 
done last fall, when there were only four Jap divisions in Burma. It could have 
been done this spring, with one or two American divisions and the cooperation 
of the Y-Force. It can still be done, with a reinforcement of U.S. ground troops 
[which Stilwell later set at a corps plus engineers], at a much smaller cost than 
is contemplated in proposed plans." 3 

As of 10 April, General Pick had only eight U.S. engineer companies 
actually working on the Ledo Road. Two engineer aviation battalions had been 
taken away from him, while the remainder of his engineers, save the eight 
companies, were working on airstrips and maintenance. 4 

The drive on Kamaing had changed in concept. Initially, on 20 March the 
112th Regiment, 38th Division, reinforced with elements of its two sister regi- 

1 The Stilwell Papers, p. 291. 

2 Rad SH 83, Stilwell to Hearn, 26 Apr 44. Item 2269, Bk 6, JWS Personal File. 

i (1) Rad CHC 1019, Stilwell to Sultan, 1 May 44. Item 153, Bk 6A, JWS Personal File. (2) 
Rad DTG 240240Z, Stilwell to Marshall, 24 May 44. Item 2740, Bk 7, JWS Personal File. (3) 
Having studied Stilwell's radios, the Asiatic Section, OPD, concluded on 22 June 1944 that since 
it could not give CBI Theater the U.S. corps plus engineers Stilwell said he would need to open 
ground communications to China, it was not in a position to direct him or Mountbatten to under- 
take operations with that in view. Therefore emphasis would be on increasing Hump tonnage, with 
ground operations secondary. Memo for Record, 6/22/2170, Case 404-2, OPD 381 Security, 

4 Ltr, Covell to Somervell, 17 Apr 44. Folder, Monthly Rpt to Somervell, OCMH. 



ments, had been ordered to envelop the 18th Division by means of a roadblock 
below Kamaing. A later order, 4 April, gave the 114th Regiment the mission 
originally set for the 112th and defined the role of the 22d Division. Its 65th 
and 66th Regiments were to hold in the Shaduzup area, while the 64th 
Regiment enveloped the 18th Division's left flank. The final directive on 23 
April, in "positive, written orders" 5 to make the record clear, called for the 
22d Division to attack, rather than hold, and to swing the 64th and 65th Regi- 
ments around General Tanaka's left (western) flank, while the 66th Regiment 
fought down the road. Once again the 112th was told to block off Kamaing 
from the south. The orders directed the "22d and 38th to be in Pakhren and 
Lawa areas by April 27. Now I've shot my wad," wrote Stilwell. Generals Sun 
and Liao at once came in to protest the orders, but to no avail. 6 

General Tanaka's mission, dating from 1943, was to hold the Kamaing area 
to the end. In March 1944 he was told by Burma Area Army that when the 53d 
Division (-) had driven the Chindits off the rail line it would be attached to 
him for a counteroffensive, providing the attack on Imphal succeeded. While 
Tanaka was waiting for the 53d Division to arrive, he was reinforced in April 
and May by two very understrength regiments, the 146th Regiment, 56th Divi- 
sion, with one battalion, and the 4th Regiment, 2d Division, with two battalions. 
Thus, at one time in May when the force from the 56th Division was attached 
to the 18th, only two regiments plus a reinforced battalion of the 56th Division 
faced the Chinese across the Salween. Elements of the 53d Division duly arrived 
in the Mogaung area, but the Japanese headquarters for north Burma, then the 
newly activated 33d Army, in Tanaka's opinion could not decide to commit 
them toward either Kamaing or Myitkyina, and his hopes for a smashing 
counterattack were lost. He therefore decided to hold above Kamaing until 
the eve of the rainy season, then withdraw to prepared positions while the 
Chinese and the Americans struggled in the floods that cover the valley during 
the rains. 7 

Though Tanaka knew that the Americans and the Chinese would have to 
take Myitkyina to open the way across north Burma, he estimated that there 
would not be a direct attack on Myitkyina as long as the bulk of the 18th was 
able to hold the Kamaing-Mogaung area. Analyzing his problem after the war, 
Tanaka remarked that after the battle of Walawbum his divisional intelligence 
section was increasingly handicapped, that he found it ever more difficult to get 
good strategic intelligence. 8 

5 Stilwell Diary, 23 Apr 44. 

6 (1) FO, 20 Mar 44, Folder, 38th Div Sitreps; FO 12, Hq Chih Hui Pu, 4 Apr 44; FO 14, Hq 
Chih Hui Pu, 23 Apr 44; FO 13, Hq Chih Hui Pu, 21 Apr 44. NCAC Files, KCRC. (2) Stilwell 
Diary, 23 Apr 44. 

7 (1) Tanaka's remarks, in Japanese Comments, Section III, are distinctly acid where 33d 
Army's action is concerned; he accuses that headquarters of endless vacillation. (2) See also his 
remarks on other occasjcfn^jn Tanaka Narrative, pages 3 and 4, and the Tanaka Interrogation in 
OCMH files. (3) Se< |Ch. IX J below. 

8 Tanaka, Japanese comments, Sec. III. 



MAP 13 

The 22d Division and the Drive on Kamaing 

When Stilwell issued orders on 23 April that the 2 2d Division was to try 
to slip its 64th and 65th Regiments around Tanaka's left, Liao's 22d Division 
was drawn up across the Kamaing Road in the vicinity of Warazup. (Map 13) 
This meant that the division was firmly established in the Mogaung valley, 
about ten miles south of the crest of the Jambu Bum. ("This war is just one 



damn valley after another!" wrote Stilwell to his wife.) On either side of the 
long narrow valley— 9,000 or so yards wide at Warazup— rose the hills now 
familiar to the soldiers. The Kamaing Road lay in the center of the valley, and 
along the road ran the stream by which nature had carved out the valley. This 
was the Nam Kawng Chaung (the upper reaches of the Mogaung River), 
whose tributaries arched out to right and left like the ribs of a fish's skeleton. 
These streams offered a series of water barriers to troops fighting their way 
down the valley; the Japanese took full advantage of them. 

When the 22d began to move south, Stilwell's orders to it were among the 
first casualties as the 22d's units attempted, not always as vigorously as Stilwell 
wished, to get around the Japanese facing them (4th Regiment to the extreme 
left of the Japanese line, across the Kamaing Road). Some of the 22d's units 
were taken out of line to refit; others simply held back. From west to east, 
General Liao had the l/65th, the 64th, the 65th, and the 66th. The 64th Regi- 
ment, 22d Division, began cutting its way around the Japanese left flank and 
occupied Hill 988, two miles west-southwest of Warazup, after the Japanese 
abandoned it. The 64th then moved southwest and reached a point well to the 
rear of the Japanese front-line positions along the Pangyu Hka. 

About this time one of the senior Chinese armored force officers took 
Colonel Brown, the tank group commander, aside and stated that both Sun 
and Liao had received orders not to advance until the Generalissimo felt it was 
safe for them to continue. The Chinese told Brown that he himself would 
continue to attack if ordered but that such conduct would seriously embarrass 
him were he (the Chinese) ever to serve under Sun or Liao. Brown reported the 
conversation to Stilwell. 

Next morning, Stilwell ordered Liao to move the 22d Division against 
Inkangahtawng, which remained in Japanese hands after Galahad withdrew. 
Liao soon moved his headquarters about a mile farther south. This move put 
the division headquarters in line with Brown's armored force command post. 
Amazed, Brown checked to see if there had been a general advance by the 
Chinese. There had been none; Liao had simply advanced his headquarters. 
Stilwell again ordered Liao to attack, and again Liao moved up his head- 
quarters, this time till it was up with that of the 66th Regiment. Almost daily, 
word would come to Brown that Liao was not going to attack until permission 
came from the Generalissimo. Brown himself thrust forward time and again 
with his tanks, without infantry support, losing several, but probably hoping 
that considerations of face would induce Liao to make whatever adjustments 
would permit him to obey Stilwell's orders and attack. 9 Presumably permission 
to attack was finally granted by the Generalissimo, possibly after he had had 
time to weigh the implications of the relief of Kohima on 20 April, for in late 
April the 2 2d Division again began to move. 

A formidable obstacle barred its way to Kamaing. Just south and east of 

9 Brown Notes, p. 14. 



TROOPS OF THE CHINESE 22D DIVISION move down the rain-soaked 
Kamaing Road. 

Nanyaseik is a range of hills, perhaps eight miles long, and rising to a 1,200- 
foot peak. There were large caves in the hills, and in them the Japanese 
emplaced four 150-mm. guns, far outranging the 155-mm. howitzers which, 
supervised by Col. George W. Sliney, were the heaviest pieces the Chinese had. 
The caves shielded the Japanese guns from bombardment and counterbattery, 
and the hills gave them excellent observation. With these pieces the Japanese 
blasted away at every profitable target that revealed itself, to as far north as 
Warazup. 10 

The 65th Regiment plus the 3 /66th, fighting down the Kamaing Road, met 
Japanese resistance north and west of Inkangahtawng. The 64th stayed behind 
the Japanese left flank, or what had been the Japanese left, until 3 May, when 
it moved again, this time straight east, across the Kamaing Road. The 64th 
Regiment cut this vital artery about five hundred yards south of the Hwelon 
Hka, four days after its patrols first crossed it. The rest of the 22d Division 
waited for clear weather to make a co-ordinated attack by air, armor, artillery, 
and the dogged Chinese infantryman on Inkangahtawng. The 4th of May met 
weather specifications and Inkangahtawng fell that day. The 64th Regiment 

10 Brown Notes, p. 15. 



promptly linked up with the rest of the 22d Division. Then the 66th Regiment 
was put in line to relieve the 65th. 11 The 22d Division had gained perhaps five 
miles down the road in these last few weeks of fighting, but the Japanese were 
"sticking it out to bitter end." Stil well's diary shows him tirelessly at work, 
going from one Chinese command post to another to urge and exhort the 
Chinese onward. 12 

The 22d Division stayed in this area for the next several weeks. During this 
time, elements of the Chinese 50th Division (149th and 150th Regiments), 
which had been flown into Maingkwan by 15 April, arrived at the front, and 
were attached to the units of the 22d Division. 

Along the Hwelon Hka, the 64th Regiment, 22d Division, was in contact 
with the Japanese. General Liao moved the 65th Regiment into line on 19 May, 
prolonging the right flank of the 64th, and the 66th Regiment was sent still 
further west. The l/66th and the 3/l49th, the latter from the Chinese 50th 
Division, plus two companies of division engineers, all attached to the 65th 
Regiment, cut a trail over the ridge marking the western boundary of the 
Mogaung valley and entered the small valley of the Lasi Hka. Moving south 
with comparative speed, they were in the vicinity of Chishidu on 27 May and 
about to cut back into the Mogaung valley. The Japanese company opposing 
them was steadily pushed back toward Nanyaseik. 

The 22d Division had thus successfully completed a maneuver which had 
failed when a task force, code name Purple, composed of the 149th Regiment, 
50th Division, plus a detachment formed from Galahad evacuees, had been 
sent on the same task. Commanded by an American, Purple Force took the 
wrong turn and, after being located by Lt. Col. Joseph W. Stilwell, Jr., in a 
liaison plane, was forced to return to the Mogaung valley. 13 On its march, 
Purple Force encountered mountains too steep for its animal transport. Two 
of the mules, loaded with medical equipment, fell over the side of the cliffs. The 
American medical personnel had to send back the pack train and carry their 
own supplies. 14 That it was an American officer who had lost the way made a 
bad impression on the l49th's commander, a very competent Chinese who may 
well have thought that he should have led the force. Thereafter, he had little 
use for Americans in any capacity. This was perhaps natural but was regrettable, 
for the 149th was an excellent regiment and its commander respected by those 

11 The 22d Division combat order of 23 April called for the 64th Regiment to swing wide 
around the 18th Division and take Nsawgatawng, just north of Kamaing. (1) G-3 Rpts, Chih Hui 
Pu, 26 Apr, 1 May, 8 May 44; Opns Rpt, 22d Div, 20 Apr 44; Daily Opns Map, 22d Div, 25 Apr 
44. NCAC Files, KCRC. (2) Combat Rpts, Fwd Ech and Misc Ln Os, 22d Div, 26 Apr 44; Situ- 
ation Maps, 22d Div, 30 Apr-2 May 44, Combat Folder, 22d Div, 23 Apr 44. Albacore Hist File, 

12 Stilwell Diary, 12 Apr 44. 

13 (1) NCAC History, II, 159- (2) Tanaka Narrative, p. 4. (3) Situation maps, 22d Div; G-3 
Rpts, Chih Hui Pu, 24 May, 30 May 44; FO 15, Hq Chih Hui Pu, 17 May 44. NCAC FHps 
KCRC. (4) Stilwell Diary, 21 May 44. (5) Col Stil well's Diary, 21 May 44. (SeelBibliographical I 
Note.) 1 1 

14 Stone MS, p. 164. 



who dealt with him. 15 The presence of the 250 or so American infantry from 
Galahad was agreed to between Stilwell and Merrill. Because of the tasks 
known to lie ahead of Galahad, Stilwell discussed the attachment with Merrill 
before ordering it. Years later, Merrill remembered telling Stilwell that 
Galahad sorely needed the 250 men and that it was really not in shape for the 
job ahead. But he was aware of the pressures and obstacles that Stilwell daily 
faced, and resolved not to add to them. 16 

Thus, the southward progress of the 2 2d Division had been painfully slow 
during May, but as the month drew to its end, the wide envelopment of the 
65th Regiment and its reinforcing battalions suggested that Tanaka would soon 
be pushed back on Kamaing itself. Stilwell and the Chinese would then be 
deep into Burma and very close to the north-south corridor through which the 
railway ran from Mandalay to Myitkyina; the corridor in turn would take them 
behind the 15 th Army. 

The 38th Division: The Generalissimo and Stilwell 

Gen. Sun Li-jen's 38th Division had the most difficult sector in which to 
operate, the hill mass that formed the eastern boundary of the Mogaung valley. 
It was in this area, incredibly difficult for troops to cross, that Galahad's stand 
at Nhpum Ga had been made a week or so before. The 38th had been moved 
into the area within which Merrill's and Hunter's men had been operating. But 
if the hills themselves were a formidable obstacle to troops, they did offer a way 
around Tanaka's right flank. And, if Tanaka moved to halt the threat posed by 
the 38th's movement south through the hills, he spread his forces thin down 
in the valley. To Stilwell, the opportunities lying before the 38th were obvious, 
so he kept constant pressure on Sun. 

After a visit to the latter's command post on 11 April, Stilwell wrote in his 
diary: "At least it looks like a start! The piled-up inertia is terrible. . . ." 17 On 
the next day the 114th Regiment relieved the 113th on the line of three villages 
all named Tingring. This move placed the 114th and 112th in line, the 112th 
to. the east forming with its lines a small salient about Nhpum Ga. The 114th 
Regiment was operating in rugged terrain which was almost as much an 
obstacle as the delaying positions directly about Kamaing defended so skillfully 
by the Japanese. On at least one occasion the 114th lost its way and had to be 
located by aerial reconnaissance. 

The 1/1 14th was to have taken the eastern Tingring and worked round the 
right flank of the 55th Regiment, but though it passed the line of the Tingrings, 
the Japanese rear guard held stoutly on Hill 1725, while the 1 8th fell back to 
the line Wala-Malakawng. Not until 20 April did the Chinese clear the hilltop. 
Stilwell chafed and fretted over the 38th Division's slowness, which he blamed 

15 Dupuy Notes. 

16 Ltr, Merrill to Ward, 26 May 52. OCMH. 

17 Stilwell Diary, 11 Apr 44. 



on General Sun. After Hill 1725 had been taken, the 114th came up against 
the Lahkraw Hka, which was in open country with little cover and dominated 
by the guns of the 18th Mountain Artillery Regiment. To the east, the 1/1 12th 
moved from Tategahtawng on 23 April to relieve End Run Force for the move 
on Myitkyina. The force was then within a mile of Manpin and created a 
salient whose tip at Manpin bulged ominously toward Kamaing and whose 
sides ran generally north of the Lahkraw Hka and the Auche-Warong trail. 18 

The slow rate of the 38th Division's advance alarmed Stilwell. After the 
First Burma Campaign he had always feared the Generalissimo might again 
try to conduct the campaign from Chungking with emphasis on the defensive. 
Stilwell surmised that again, as in 1942, the Generalissimo was corresponding 
directly with Sun and Liao. On 15 April Stilwell received a radio from the 
Generalissimo observing that the Mogaung valley was "good for attacking and 
defending," so Stilwell should be careful. Stilwell promptly called the matter 
to Marshall's attention. He told the Army's Chief of Staff that in his opinion 
the hard core of Japanese resistance had been broken, that the Chinese with 
little further trouble could go rapidly on to Kamaing. Stilwell concluded that 
Sun's and Liao's conduct of operations could only be explained by secret orders 
from the Generalissimo to slow their pace. He asked Marshall to keep the 
President informed of this situation. 19 To placate the Generalissimo, Stilwell 
had Hearn present an optimistic appraisal of the Burma campaign with the 
tactful statement that Stilwell would move cautiously by exploiting every 
Japanese blunder. 20 

The 22d and 38th Divisions continued to plod along. Completely appre- 
hensive, Stilwell asked General Liao bluntly if the Generalissimo had ordered 
a slowdown. Liao replied that the Generalissimo did correspond with him 
directly and assured Stilwell that the Generalissimo had ordered him to obey, 
even when Stilwell was wrong. Far from reassured, Stilwell asked Hearn to 
raise the issue with Madame Chiang, specifically, to ask whether the General- 
issimo had sent a message encouraging Sun and Liao as Stilwell had requested. 21 

After the war, an officer of Sun's staff happened to meet with Col. Thomas 
F. Van Natta, III, who had been the senior liaison officer with the 38th Divi- 
sion. They refought the campaign, and when the discussion turned to the long 
delay above Kamaing, the Chinese stated that he personally had seen several 
messages from the Generalissimo to Sun, as of this period, ordering him to 
proceed with caution. 22 

18 (1) Lt. Col. Thomas F. Van Natta, III, History of the 38th Div, CAI, 1 Oct 43-31 Aug 45, 
MS. (Hereafter, Van Natta MS.) NCAC Files, KCRC. (2) G-3 Rpt, Hq Chih Hui Pu, 26 Apr 44. 
NCAC Files, KCRC. (3) Tanaka Narrative, App. (Overlay) II. (4) Stilwell Diary, Apr 44, 
2 May 44. 

19 (1) Memo, Generalissimo to Hearn for Stilwell, 6 Apr 44. Item 2219, Bk 6, JWS Personal 
File. (2) Rad SH 34, Stilwell to Marshall, 16 Apr 44. SNF 131. 

20 Memo, Hearn for Generalissimo, 16 Apr 44. Item 2221, Bk 6, JWS Personal File. 

21 (1) The Stilwell Papers, p. 290. (2) Rad SH 45, Stilwell to Hearn, 17 Apr 44. Item 2231, 
Bk 6, JWS Personal File. 

22 Van Natta Notes. 



The Generalissimo's response to Hearn's and StilwelPs efforts was 

I am in receipt of reports through the kindness of General Hearn. It is gratifying to know 
that if the situation in the Kohima area can be stabilized, the problem of supply and morale 
of the units now fighting in northern Burma will not deteriorate. 

I beg to express my deepest respect to you for your leadership and direction of the 
progress we are making in northern Burma. Although I hold no responsibility in the com- 
bined operation of the present Burma campaign, I hope we will not cause unexpected 
damage and failure to our friends and units. That's why I am so concerned with the general 
situation in the Salween district. 

We will exert to the fullest in doing whatever necessary to the attainment of our victory. 
Please do not unduly worry. 

The Chinese Expeditionary Force in the Salween district is now prepared for counter- 
offense, and is very attentive in watching for the most opportune moment to coordinate with 
the campaign for Myitkyina. It is impossible now for us to directly reinforce northern 
Burma. But I hope you will maintain the limited offensive strength now at your disposal, 
and be judicious in its employment. 

Generals Sun and Lieu [sic], under your brilliant leadership, have done valuable services 
for which I have extended to them my personal congratulations. 

Respectfully yours, 


The passage that struck Stilwell with greatest force was the Generalissimo's 
assurance that he would keep his friends and troops from undue losses. Stilwell 
promptly relayed the message to Marshall. Analyzing his situation for the Chief 
of Staff, Stilwell concluded that he could take no remedial action, for any new 
division commanders would also get secret orders from the Generalissimo, and 
that anything precipitate would "risk the loss of all the results we have gotten 
to date." In effect, Stilwell would have to persevere in his exhortations and 
hope that they would be enough to bring the Chinese into Mogaung and 
Myitkyina. 24 

So the 22d and 38th Divisions continued as before. When on 26 April 
General Sun came in with his plan to take Kamaing, Stilwell disapproved it, 
"verbally, and in writing." Sun came back next day with another, and that, too, 
Stilwell rejected. The 38th Division went ahead on the basis of existing orders. 25 

Turning Tanaka's Flank 

To break the stalemate along the Lahkraw Hka, the 114th Regiment had to 
clear away the Japanese observation posts in the hills. The l/ll4th on the 
regiment's east flank cut around the flank of the 55th Regiment on 28 April, 

" Ltr 654, Generalissimo to Stilwell, 17 Apr 44. Item 2252, Bk 6, JWS Personal File. 
24 Rad SH 73, Stilwell to Marshall, 24 Apr 44. SNF 131. 
" Stilwell Diary, 26, 27 Apr 44. 



while the 2/ll4th pushed the same Japanese unit back a few hundred yards. 
This bending process continued during the next two days, and the 114th Regi- 
ment was well south of the enveloped Japanese flank and within half a mile of 
Wala. The 1st and 2d Battalions of the 114th then began moving straight 
south, leaving behind them pockets of determined Japanese who held up the 

The 113th Regiment came back into line at the end of April to put frontal 
pressure on the Japanese positions along the Lahkraw Hka and Tigrawm Hka. 
Meanwhile, the 112th Regiment on the 38th Division's east flank was holding 
its salient without attempting to move. The Japanese managed to stabilize 
their lines on the creeks east and west of Wala and hold there until 6 May when 
a company of the 114th crossed the Nawngmi Hka and started the advance 
going again. Two days later the ll4th's advance became general, while on the 
east at Stilwell's order the 1 12th began to gather its companies to take Warong. 26 

East Wala and Hlagyi were taken on 9 May, and the ll4th's penetration, 
driving deeper into the Japanese lines, began to approach the 112th's outposts 
just north of Manpin. By 12 May the 114th and 112th Regiments were able to 
maintain communications, with very few Japanese between them. The ll4th's 
penetration weakened the position of the Japanese facing the 113th, and the 
113th's line began to roll up slowly from east to west as the 1st and 2d Battal- 
ions went through what had been the center of tht55tb's and 56th's positions 
to press on to Wala and Maran. 27 

The whole right flank of General Tanaka's position in the Mogaung valley 
was now crumbling. To bolster it, Tanaka ordered the I /55th, which had been 
his extreme right (east) flank detachment (and significantly, regarding his 
casualties, was commanded by a captain) to be reinforced by the 146th Regiment, 
56th Division. Command of this task force was given to Maj. Gen. Toshiji Aida, 
Tanaka's infantry group commander. 28 About this time, the officers of the 18th 
Division learned that there was no chance of the Japanese counteroffensive in 
the Hukawng Valley on which so many hopes had been staked, for the U oper- 
ation was failing "miserably." Their morale began to sag, and in Tanaka's later 
opinion this realization marked the "turning point" for his division. 29 

The Japanese held desperately at West Wala, Maran, and Sharaw, tempo- 
rarily stalemating the 113th Regiment. The 114th was sent hooking west 
behind these Japanese strongpoints, and the two regiments took them one 
after another between the 18th and 25th of May. To the east, in front of the 
112th Regiment, the Warong position fell on 20 May. It had been stubbornly 
defended, and a number of Japanese were cut off along the Auche- Warong 
trail north of there. With the clearing of the Japanese from these strongpoints 

26 (1) Van Natta MS. (2) 38ch Div Sicreps, 29 Apr-8 May 44. NCAC Files, KCRC. (3) 
Stilwell Diary, 5 May 44. 

27 3 8th Div Sitreps, 8-19 May 44. NCAC Files, KCRC. 

28 Tanaka Narrative, p. 4. 

29 Tanaka, Japanese Comments, Sec. III. 



on the east side of the Mogaung valley, the right side of Tanaka's line seemed 
to be dissolving, even though the Japanese were still falling back slowly before 
the 22d Division on the center and left. 30 According to Tanaka, his right was 
in "a state of confusion." Part of the 55th Regiment still clung to a little piece of 
the Wala heights, with Chinese infantry between the regimental headquarters 
and the battalion position, and other Chinese to the rear of regimental 
headquarters. 31 

The 112th Regiment's Stand at the Seton Block 

On the night of 19 May, General Sun probably received the Generalissimo's 
permission to take Kamaing, for the next morning at breakfast he told the 
American liaison officer, Colonel Van Natta: "We go on to take Kamaing 
now." Promptly, Sun and Van Natta called on Stilwell with Sun's plan to take 
the town. The change in Sun's attitude "astonished" Van Natta. Previously, 
Sun had been all difficulties, delays, complaints, and objections. Now, though 
Stilwell could promise to airdrop only 50 percent of the supplies he needed, Sun 
raised no objection. 32 Sun's plan was simple and direct: send the 112th Regi- 
ment wide around the Japanese east flank to cut the Kamaing Road south of 
Seton. The 114th would exploit the evident confusion and weakness of the 
Japanese on the east by moving steadily down the Kumon Range to 
Tumbonghka near the point at which the Mogaung valley merges with the 
Irrawaddy valley. Its goal was Mogaung. This plan Stilwell approved. 33 

On the 22d Division front, the task force which had taken Chishidu kept 
briskly toward the Kamaing Road several miles south of the Japanese position 
at Mataing Sakan. The 2 2d Division was thus keeping Tanaka fully occupied. 
Meanwhile, to the east, the 112th was cutting its way through the jungle, 
unobserved by the Japanese. The commander of the Japanese right flank, Aida, 
was watching events to the direct north, and the 112th slipped around him. 34 

On 25 May the 112th (-) emerged from the jungle well in the rear of 
Tanaka's lines, and crossing the Mogaung River, burst into one of Tanaka's 
major supply centers. The Chinese captured 35 Japanese trucks, a jeep, a sedan, 
8 warehouses of food and ammunition, 100 horses, 4 pieces of artillery, a work- 
shop, and a motor pool. It was a great coup for General Sun and the Chinese, 
and a major crisis for Tanaka. If the 112th could keep its grip on the Kamaing 
Road, the 18th Division would be in its last battle. General Aida, who had had 

30 The Japanese were feeling the drain of the last six months' fighting. The 6th Company, 
2/ 56th, was down to fifty men as of 15 April. ( 1) Statement, Superior Pvt Fujiyoshi Kawaguchi, 
2 Oct 44. Folder 62J 41-50, NCAC Files, KCRC. (2) Van Natta MS. (3) 38th Div Sitreps, 15-22 
May 44. NCAC Files, KCRC. 

31 Tanaka, Japanese Comments, Sec. III. 

32 Van Natta Notes. 

33 ( 1) The Stilwell Papers, p. 297. (2) G-3 Rpts, Hq Chih Hui Pu, 30 May, 10, 13 Jun 44. 
NCAC Files, KCRC. 

34 Tanaka, Japanese Comments, Sec. III. 



strict orders from Tanaka to hold the village of Lawa, where several trails con- 
verge east of Kamaing, abandoned his position without authority, and the 
whole of Tanaka's right rear flank was wide open. 35 

Tanaka stripped his division headquarters and service units of available 
troops and attacked the 112th at once. From his left flank he called in the 4th 
Regiment, 2d Division, under Col. Yusaku Ichikari to attack the Chinese from 
the north. On the south, the 18th Division's service troops were hastily assem- 
bled and hurled at the block. Simultaneously, Tanaka ordered his chief engineer 
to cut a secret escape route. With great relief the Japanese learned that such a 
route had already been reconnoitered. The trail ran from the Noidaw Bum to 
the Bumrawng Bum via Noidawyang. 36 

The Japanese battered in vain at the Seton Block. The Chinese fought 
valiantly. Rain, floods, and constant Japanese attacks made the action one of 
the most trying of the campaign; the Chinese stood up to it with fortitude and 
devotion. Though Tanaka was withdrawing his forward elements from their 
positions in the Mogaung valley, it was probably obvious to him that complete 
withdrawal from the Kamaing area would find the 18th minus its artillery and 
vehicles and so weakened as to need complete rebuilding. Moreover, he had 
orders to defend Kamaing to the last. Therefore, he had every incentive to 
break open the line of communications to the south, and his men did their best. 
Attack after attack was hurled at the 112th, but the Japanese on both sides of 
the Seton Block were suffering from malnutrition and disease; many of those 
to the south were replacements, and the 112th held doggedly. At the end of its 
ordeal, only two of the 112th's officers were on their feet. 37 

The success of the 112th Regiment in holding the Seton Block suggested 
that the climax of the valley campaign was close at hand. The Chinese, now 
that t he General issimo had lifted all restrictions, drove in on Tanaka from all 

sides. (Map 14) The 2 2d Division task force which had been moving toward 
the Kamaing Road moved squarely across it on 1 June, cutting off the 55th, 
56th, and 4th Infantry Regiments from the withdrawal route in the hills by which 
Tanaka had planned to bypass Kamaing. Such was the current disorganization 
of the Japanese that the commander of the 55th Infantry Regiment was in con- 
tact with but one company of his regiment. The 2d Battalion (-) of the 56th, 
supported by six 75-mm. and two 150-mm. pieces, tried to force the 22d Divi- 
sion off the Kamaing Road, but failed. Since the Chinese barred the escape 
route via the Noidaw Bum, the 56th had to make a crude trail to the southeast, 
from the Pakhren Bum area, in the hope that, covering some 500 sick, the 
artillery, and the motor transport, it might swing east around the 22d Division's 
roadblock, and then back onto the escape trail. 38 

35 Tanaka, Japanese Comments, Sec. III. 

36 (1) G-3 Rpts, Hq Chih Hui Pu, 30 May, 10 Jun 44. NCAC Files, KCRC. (2) Tanaka 
Narrative, pp. 4-5. 

37 (1) Tanaka Narrative, p. 6. (2) Van Natta Notes. 

38 (1) Tanaka Narrative. (2) Tanaka, Japanese Comments, Sec. III. 



With the 22d Division blocking him just southeast of Nanyaseik, and the 
38th firmly holding the Seton Block, Tanaka was faced with the gravest tactical 
problems. Nevertheless his skillful delaying action, greatly aided from Decem- 
ber to May by Chinese tactics that to the Americans appeared willful delay, was 
yielding results for the Japanese. His attempt to hold until the monsoon closed 
down on the Chinese and on the American SOS showed acute appreciation of 
the factors in his favor. By June about one inch of rain fell daily. Though 
supply convoys could still move from Ledo to Shingbwiyang, the combat trail 
from Shingbwiyang south was very difficult. The road was graded to Tingkawk 
Sakan (mile 164) and metaled almost to Mile Post 138. However, rainfall had 
blocked the road over the flats north of Tingkawk Sakan. At the end of June 
the situation was unchanged. The Japanese stand at Kamaing and the heavy 
rains immobilized the survey party and the road trace. 39 As the flood waters 
rose in the valley, they effectively barred armor from moving south to Mogaung 
or Myitkyina. 40 

Defeat of the 18th Division 

The condition of the 18th Division was now almost desperate. The rice 
ration, normally 860 grams per day, had shrunk to 100 grams. Lack of gasoline 
immobilized the Japanese trucks. Allied bombers had destroyed about 40 per- 
cent of the supply dumps. The artillery was rationed to four shells a day. The 
18th's units were far understrength. Where in April they had had perhaps fifty 
men to a company, now they were down to thirty. 

The 56th Regiment in particular was in grave straits. Its commander reported 
on 7 June: 

The advance attack of the enemy from the north is unexpectedly swift; the enemy is 
advancing southward, threading through the gaps in our lines by wading chest-high through 
marshy zones. I am unable to contact the 1 and 3 Battalions, which are under my command, 
and their situation is unknown. The platoon occupying the vicinity of Nanyaseik received 
an enemy onslaught and all troops were annihilated. The enemy stormed into our main 
artillery position, and with our motor trucks, artillery and other vehicles crowded together 
in the vicinity of the narrow, forked road, there is much confusion. The transfer of most of 
the patients has been completed. The regiment will cover the withdrawal of the main body 
of the division at the sacrifice of our lives. I believe this will be our final parting. Please give 
my best regards to the division commander. 

The survivors of the 56th's infantry made good their escape to the south- 
east, then back to the sheltering hills, but the artillerymen died alongside their 
six 75-mm. and two 150-mm. pieces as the Chinese 149th Regiment overran 
them, while the motor transport was destroyed. The colors of the 56th Regiment 
were safely conveyed to division headquarters in Kamaing. 41 It was victory and 
complete evidence of the high martial qualities of the Chinese soldier. 

3 9 SOS in CBI, pp. 451-52. 

40 Brown Notes, p. 15. 

41 (1) Tanaka Narrative, pp. 5, 6. (2) Dupuy Comments. (3) Tanaka, Japanese Comments, 
Sec. III. 



The 113th Regiment in its turn moved south down the foothills of the 
Mogaung valley's east side. It took Lawa, then Zigyun, directly across the Nam 
Kawng Chaung from Kamaing. The 113th began to probe for a crossing. To 
the north and west of Kamaing, the 22d Division and the 149th Regiment 
kept the Japanese under heavy pressure. As the Japanese began to shift their 
forces west toward an escape route, they left a gap in their positions through 
which the 3d Battalion of the 65th Regiment made a deep penetration, permit- 
ting a wide envelopment that placed the battalion south and west of Kamaing. 
The remainder of the 65th poured on through. Meanwhile, the 149th made a 
close-in envelopment of Kamaing. Kamaing was now encircled to the west, 
north, and east, with the 149th due west. On 16 June the 149th moved out of 
the bush, across the fields, and into Kamaing to take the settlement. With 
Kamaing in Chinese hands, the 18th Division was pushed into the hills south 
and west of Kamaing. 42 

On that same 16 June that Kamaing fell, the 114th Regiment which had 
been sent south past the fighting around Kamaing met the Chindits at Gurk- 
haywa. Their meeting established a ground line of communications to the 
Chindits and meant that the Allies were solidly established in Burma, just about 
two years after the end of the First Burma Campaign. 43 

The experiences of the surgical team from the 43d Portable Surgical Hos- 
pital which accompanied the 114th Regiment illustrate the marching qualities 
of the Chinese Army. By 28 May the terrain over which the 114th was making 
its way was so bad that horses could not be used. The medics then took with 
them only 100 pounds of supplies, carried by the Chinese. Unhappily, some of 
the medics' scanty food stock was stolen, and their rations were exhausted 
around 30 May. The inhabitants of the little village of Kawnan contributed 
some pork and rice. From 1 to 4 June the Americans lived on a cup of rice and 
water twice a day, which they obtained from the Chinese. Their operating table 
was a litter on two bamboo trestles over which hung a little thatched roof. 
Under these conditions, the surgical team performed 138 operations with but 
three postoperative deaths. When they rejoined the parent unit on 16 June, the 
members of the team had lost twenty-five pounds each on the average, and 
staggered as they walked. The 114th Regiment was still battleworthy. 44 

On 22 June, Headquarters, 33d Army, ordered Tanaka to withdraw from the 
Kamaing area to the area north of Sahmaw. Tanaka argued that he should stand 
his ground while the 53d Division, which had done so little to restore the Japa- 
nese fortunes, pushed aside the Seton Block and reopened his line of communi- 
cations. The 33d Army agreed, and for three days Tanaka and the 53d Division 
tried to pry the Chinese away from their strangling hold. But the 112th 

42 (1) Dupuy Comments. (2) G-3 Rpt, Hq Chih Hui Pu, 10, 13 Jun 44. NCAC Files, KCRC. 

43 Stilwell Diary, 16 Jun 44. 

44 Stone MS, pp. 169-74. 



Regiment was now reinforced by the 113th, and they were too strong for the 

Tanaka reported as much, and 33d Army ordered the 18th Division to with- 
draw to the hills north of Sahmaw, a village on the Burma railway south of 
Mogaung. While the 4th and 146th Regiments made covering attacks, the 
remnants of the 33th and 36th Regiments destroyed their artillery and heavy 
equipment, then withdrew along the escape trail cut through the forest west of 
the Seton roadblock, the same trail of which Tanaka had earlier learned with 
such relief. The Japanese rear guard fell in behind the ragged survivors of the 
elite division that had taken Singapore in 1942, and the starving, malaria-ridden 
Japanese slowly trudged out of the Mogaung valley. The Chinese Army in 
India, with vital assistance from the 3d Indian Division and Galahad, had 
cleared the Japanese from the Hukawng and Mogaung valleys. 

By keeping intact the blockade of China for yet another year, the 18th 
Division and Tanaka may have profoundly affected the history of Asia. If Stil- 
well had won a speedy victory in north Burma, the position of the Generalis- 
simo's government in China could have been greatly strengthened by the return 
of good Chinese troops and the delivery of trucks and artillery to China in 1944. 
But events did not fall that way. 

The Japanese paid heavily to hold north Burma and prolong the blockade 
of China. The 18th Division lost 50 percent of the strength with which it began 
the action; the 4th and 146th Regiments, perhaps 33 percent. Of the 18th Divi- 
sion, 3,000 men succeeded in withdrawing; of the two regiments, about 1,000 
men each. 45 

Stilwell and the Chindits 

The meeting between Chinese and Briton, lao ping and Tommy, on 16 June 
dramatized the work of the 3d Indian Division in cutting the Japanese line of 
communications to Kamaing and Myitkyina and in containing substantial 
Japanese forces. The role of the Chindits occasioned heated controversy at the 
time, and some of the embers were still glowing after the war. A detailed 
account is beyond the scope of this volume, for the exchanges were numerous 
and bitter. The withdrawal of the Chindits from Mawlu and the belief, which 
the Chindits shared with the men of Galahad, that "after 90 days we get out 
of this epithet place," a belief in which SEAC concurred, led to strong differ- 
ences between Stilwell and his staff on one side and Lentaigne, Mountbatten, 
and SEAC on the other. 46 

45 (1) Tanaka Narrative, p. 7. (2) Tanaka Interrog. OCMH. 

46 Stilwell Numbered Files 4 and 172 plus Folder, Memos and Operational Instructions (CAI, 
NCAC, 3d Indian Division, January-May 1944, NCAC Files, KCRC), contain many of the records 
in possession of American officers or agencies. The SEAC War Diary and Admiral Mountbatten's 
report give another point of view. 



The officers of the 3d Indian Division and SEAC held as doctrine that when 
soldiers had been ninety days behind the Japanese lines they were ipso facto 
exhausted and should be withdrawn from the field. The belief was Wingate's, 
drawn from his 1943 experience, and was known to the men in the ranks. 
Moreover, the monsoon rains made the whole process of air supply most diffi- 
cult, while the waters made cross-country marches extremely trying. Stilwell 
objected that the simple presence behind Japanese lines of the Chindits was no 
support to him, that they aided his operations only when they actually placed 
themselves across the Japanese lines of communications. Stilwell feared that in 
retreating to the north the Chindits would bring with them the swarms of 
Japanese they had attracted. He also feared the effect on the morale of Chinese 
who had been engaged since November 1943 if troops who entered the field 
only in March 1944 passed through the Chinese lines on their way out. 

On 17 May, 3d Indian Division came under Stil well's operational control, 
and Lentaigne placed his headquarters next to Stilwell's. Stilwell did not desire 
operational control, for he did not believe his orders would be accepted. More- 
over, he was under constant pressure from SEAC and the Chindit commanders 
to agree to their early evacuation. While Stilwell and Lentaigne were conferring 
on 25 May on holding a position (Blackpool) near the railway in the vicinity 
of Namkwin, and were agreeing that it should be evacuated only in case of 
emergency, the block was being evacuated. Stilwell's anger at this course of 
events, together with the steadily declining strength of the Chindits, created a 
crisis which soon required the attention of the highest SEAC officers. 

Then, Stilwell ordered the 77th Brigade of the Chindits to attack Mogaung. 
Its capture would be a great aid to Stilwell's Chinese and would pull out the 
keystone of the whole Japanese position in north Burma. Between 22 and 26 
June, 77th Brigade attacked Mogaung with a dash and gallantry that drew 
praise from U.S. observers, but it lacked the weight to overrun Mogaung in 
one rush. The 114th Regiment, 38th Division, joined in the fight. Mogaung 
fell on 26 June and both units claimed credit for the victory. 

The commander of the 77th Brigade now insisted that his men be with- 
drawn as unfit for further service and soon after himself ordered them 
withdrawn. Charges of bad faith were freely made on all sides and feeling ran 
high. Stilwell believed the 77th Brigade had been withdrawn in disobedience 
of written orders to hold its ground and prevent the Japanese from sending 
heavy reinforcements to Myitkyina. Lentaigne assumed responsibility for the 
act as necessary to preserve his troops. 

There was still another clash between Stilwell and the Chindit headquarters 
in June, when Stilwell asked Lentaigne to report why the 111th Brigade had 
not complied with a series of orders issued between 8 and 17 June. These called 
for the 111th to occupy positions in the rather large area between Pahok and 
Sahmaw in order to guard the right, or western, flank of the 3d Indian Division. 
The matter was aired at a meeting on 30 June between Mountbatten, Stilwell, 



Lentaigne, and members of Stilwell's staff. There it developed that Stilwell 
had misunderstood the orders his staff had issued the 111th. These had called 
for the occupation of an area, rather than of a point, Sahmaw itself, as Stilwell 
had believed. The conference agreed on a further evacuation of the Chindits 
after the southward advance of the Allies in north Burma had reached Sahmaw. 

In the weeks that remained of the Chindits' campaigning the 111th Brigade 
briefly seized Point 2171, a hill feature north of Taungni, which was a critical 
terrain feature controlling approaches to the railway. The Japanese defended 
it stubbornly; significant of the bitterness of the fighting was the posthumous 
Victoria Cross awarded Maj. Frank G. Blaker. The 14th Brigade relieved the 
111th and, reoccupying Point 2171 in the process, cleared the hills to the 
western side of the railway. 

Fortunately, the veteran British 36th Division was arriving in the forward 
area to replace the battered Chindits; the steady success of the campaign in 
north Burma made air space for evacuation available, and sober second thought 
prevailed. In retrospect, it seems apparent that neither Stilwell nor Mountbatten 
wanted to let so potentially grave a dispute develop further. After heated dis- 
cussion, the charges made were simply dropped, and the flood of events began 
to sweep over the episode, as the jungle closed over the paths and clearings 
where the Chindits had fought and died. 

As the Chindits were slowly coming out of Burma there was apparently 
some disposition to disparage their accomplishments. SEAC considered they 
had yielded 5 percent return instead of the hoped-for 15 percent— which latter 
would have been generous indeed. Some considered that Galahad and 23d 
Brigade (which had fought the 15th Army), with their tactic of close-in envel- 
opment, had demonstrated the correct use of long-range penetration groups. 47 
A minute of Giffard's based on the view that Galahad and 23d Brigade had 
been correctly used suggested forming six long-range penetration brigades for 
the next campaign. When that campaign began, Fourteenth Army in its supply 
arrangements and its tactics reflected many of the pioneering Chindits' 

In assaying what Wingate and the Chindits had done, a group of former 
Japanese officers wrote: 

The Chindits interfered with the Imphal Operations from the very start and forced 15th 
Army to divert one battalion each of the 1 5 th and 33d Divisions, to deal with them. Also 
diverted was the main force of the 53d Division which was to be the general reserve for the 
Burma Area Army and was, if there had been no such emergency as the descent of the 
Chindits, to have reinforced the 15th Army at Imphal. The 5th Air Division was obliged to 
use half of its strength against the Chindits when its full strength should have been employed 
to support the 15th Army. 48 

47 (1) Telg 2739, Air Ministry to SACSEA, COS (RL) 206, 25 May 44; SAC (44) 307, 30 Jun 
44, sub: Reconstruction of LRPG's by CinC, Eleventh Army Gp. SEAC War Diary. (2) Lt. Col. 
J. E. B. Barton, Special Force in Direct Support of C.A.I. , draft monograph. Cabinet Office, Hist 
Sec, London. 

48 Japanese Officers' Comments, p. 18. 



As for the operations of the 18th Division, Japanese sources state that cutting 
its supply line made its holding operation useless. 49 

The March to Myitkyina 

The clearing of the Mogaung valley accomplished one half of Stilwell's 
objective for the spring of 1944 and brought nearer the ultimate accomplish- 
ment of the other— the taking of Myitkyina, the goal of this campaign. His 
aim, indeed, had been to take Myitkyina first, but events had not fallen that 
way. On 21 April Stilwell set up a force to seize Myitkyina, and called it End 
Run (harking back to his college football days, as he often liked to do). 
Wasting away as rapidly as were the Chindits, Galahad now had but 1,400 
men of its original 2,997, and so they had to be combined with Chinese troops. 

Three combat teams were created: H Force under Colonel Hunter (150th 
Regiment, 50th Division, and the 1st Battalion of Galahad; 3d Company, 
Animal Transport Regiment, and a battery of the 22d Division artillery); 
K Force under Colonel Kinnison (88th Regiment, 30th Division, and 3d Bat- 
talion of Galahad); M Force under Colonel McGee (the 2d Battalion of 
Galahad plus 300 Kachins). With them went surgical teams from the Sea- 
grave Hospital Unit and from the 73d Evacuation Hospital, plus the whole of 
the 42d Portable Surgical Hospital. Whatever prestige might accrue from the 
swift seizure of Myitkyina might have been the Generalissimo's, for Stilwell 
now urged him to send a Chinese division from China to Myitkyina via the 
Hpimaw pass. The Generalissimo refused. 50 

Shortly after the siege of Nhpum Ga, Merrill from his sickbed sent a staff 
officer to Hunter's headquarters to advise him that a thrust over the mountains 
at Myitkyina was contemplated. This was a change from the original plan for 
working down the Hukawng and Mogaung valleys, and terrain data plus staff 
studies were obviously called for. Hunter at once put the Galahad staff to 
work. From their studies they prepared a rough plan which was returned to 
Merrill. Their plan appeared an improvement o ver Merril l's original ideas and 

he drafted the final plan for Stilwell accordingly. (Map 15) 

The version approved by Stilwell called for crossing the Kumon Range 
with H and K Forces via the Naura Hkyet (a 6,1 00- foot pass), then turning 
south on Ritpong. From there the two forces would take separate routes that 
would later converge on Myitkyina. M Force would be in position to cover the 
south flank, the most likely danger spot, though in such a march through 
territory nominally in Japanese possession no one could guess from what 
quarter the enemy might attack, and which force would actually have the 
blocking role. 

49 SEATIC Bull 240, p. 9; SEATIC Bull 242, p. 20; SEATIC Bull 247, pp. 12-13. MID Library. 

50 (1) FO 14, Hq Chih Hui Pu, 21 Apr 44. NCAC Files, KCRC. (2) Galahad, pp. 87-88. 
(3) Stone MS, p. 182. (4) Rad SH 36, 16 Apr 44. Item 2228, Bk 6, JWS Personal File. 




Merrill, somewhat improved in health, again assumed command of the 
force, with orders to take and hold the Myitkyina airstrips. In giving Merrill 
his orders for the march, Stilwell stated that he knew he was calling on Gala- 
had for more effort than could fairly be expected, but that he had no other 
option. In the light of that, and the exhaustion of the unit, he authorized 
Merrill to begin evacuating Galahad "without further order if everything 
worked out as expected." 51 

After discussing the plan with his battalion commanders, Merrill said a few 
words about what would be done for Galahad on completion of the mis- 
sion. 52 Such a prospect was a tremendous incentive to the weary men of 
Galahad, and Merrill believed many made the march as a last desperate effort 
for a great prize. 53 Capt. William LafHn was sent ahead with 2d Lt. Paul A. 
Dunlap, thirty Kachin Rangers, and thirty coolies to make the trail negotiable. 

K Force moved out on 28 April; H Force, on 30 April. The so-called trail 
over the pass was more nearly a route used by the Kachins; in some places 
there was no path. Twenty pack animals slipped and fell off the narrow, muddy 
way. It was a grueling march for men who had already marched 500 miles and 
fought several battles, most of the time on K ration. Before End Run Force 
reached Ritpong, it had only one contact with the Japanese. A few of the enemy 
were flushed from cover by the l/88th, but it was feared the Japanese held 
Ritpong in strength. 54 

On 5 May, when the leading elements of K Force (3d Battalion of 
Galahad and the 88th Regiment, 30th Division) were a mile from Ritpong, 
Colonel Kinnison began an envelopment to hit the village from north and 
south. The 3d Battalion managed to cut its way through the woods and place 
itself across the southern approaches to Ritpong on 6 May. Meanwhile, the 
89th Regiment tried to take Ritpong from the north but failed. Next day an 
American reinforced company attacked Ritpong from the south but was stalled 
by a machine gun nest. 

Since the Chinese were attacking skillfully and well, Colonel Kinnison 
decided to let them take Ritpong, while the Americans blocked the trail to the 
south. During the night of 7-8 May Japanese breakout attempts were repulsed. 
The Chinese attacked again on the 8th, and Ritpong fell on the 9th. The delay 
at Ritpong permitted H Force to overtake and pass through K Force. 

Meanwhile, M Force (2d Battalion of Galahad, plus 300 Kachins) had 
had one skirmish with Japanese, routed them, and begun to cut its own paths. 
The march was fatiguing in the extreme; fully half the animals died of exhaus- 

31 (1) Quotation is from Merrill Notes. (2) Hunter Notes. 

52 Ltr and Rpt, Office IG, CBI, to CG, USAF CBI, 12 Jul 44, sub: Investigation re 5307th 
(Prov) Unit (Galahad) (hereafter, Galahad Investigation), pp. 10, 17, 21, 64. AG 333, NCAC 
Files, KCRC. 

53 (1) Merrill Notes. (2) Hunter Notes. 

54 Galahad, pp. 87-91. 



tion or fell into the gorges. The men were farther harassed by fevers and 
dysentery, but they were only two days behind H and K Forces. 55 

Proceeding from Ritpong, K Force feinted toward Nsopzup, a Japanese 
supply point on the trail from Myitkyina to Sumprabum, to attract the Japa- 
nese. While so engaged, K met a Japanese force near Tingkrukawng that was 
strong enough to pin both of its combat teams to the ground and then to halt 
the Chinese when they were committed. Attempts to envelop the Japanese 
failed. Since H Force (1st Battalion of Galahad and the 150th Regiment, 50th 
Division) was proceeding unmolested, Merrill told Kinnison to withdraw. 
K Force then picked up H's trail and followed it to Myitkyina. Moving on, 
Kinnison reached Hkumchet In on 17 May and paused to await an airdrop. 
M Force was similarly engaged at Arang. During these strenuous days, Kinni- 
son became ill of the deadly mite typhus, was evacuated, and died with shocking 
speed. In all, 149 men contracted the little-known scourge. Many of them died. 56 


While K Force was fighting at Tingkrukawng, H Force kept on to a river 
just south of the village of Namkwi on the Mogaung-Myitkyina railroad and 
about two miles from the principal Myitkyina airstrip that lay almost due west 
of Myitkyina itself. Despite the several brushes with the Japanese, Kachin 
informants were sure that the Myitkyina garrison was not on the alert. To 
ensure surprise, before Colonel Hunter and his force bivouacked for the night 
of 16 May they rounded up the local Burmans and kept them under careful 
watch. A patrol reported the airstrip lightly held. Hunter decided to attack at 
1000, 17 May. The 150th Regiment was ordered to take the airstrip, while 
Galahad personnel took the Irrawaddy ferry terminal at Pamati. The other 
airstrip, north of the town, was left alone for the present. 

The attack went like a service school demonstration, for though the Japanese 
knew Myitkyina was in danger, the actual assault was a complete surprise. 
Colonel Maruyama, the Japanese commander, had two understrength battalions 
of the 114th Regiment in the town of Myitkyina and in its little suburb of 
Sitapur. There were 100 more men of the 15th Airfield Battalion on the north 
and south airstrips, 318 men from labor and service units on various details in 
Myitkyina, and 320 patients in a military hospital. Perhaps 700 able-bodied 
Japanese were present when the battle began. 57 

As soon as Colonel Hunter considered his hold on the major airstrip secure, 
he sent the prearranged code signal, Merchant of Venice, which meant the 
process of supply and reinforcement could begin. Previous code signals sent 

" Galahad, pp. 91-97. 

56 Ltr, Brig Gen Waldemar F. Breidster to Ward, 4 Sep 51. OCMH. 

57 (1) Galahad, pp. 97-101. (2) G-2 Per Rpt, Hq Chih Hui Pu, 24 May 44; PW History, 
Folder, Current Chronological PW File. (Hereafter, PW History.) NCAC Files, KCRC. 



A BRITISH ANTIAIRCRAFT UNIT at Myitkyina airfield. 

forty-eight and twenty-four hours in advance of the actual descent on the air- 
strip had alerted Stilwell's headquarters, and the process of reinforcement had 
been prepared. It had been feared that the Japanese might have been able to 
damage the airstrip. So the Air Forces had gliders with aviation engineers 
standing by. Hunter reported by radio that the strip was in good condition, 
word which by agreement with Merrill was to have begun a flow of food, 
ammunition, and infantry. 58 

The process of reinforcement was a disappointment to the waiting troops. 
A company of the 879th Engineer Aviation Battalion came in via glider; a 
battery of .50-caliber antiaircraft was flown in; and then the 2d Battalion, 
Chinese 89th Regiment, arrived before the weather closed in on the 17th. 59 

At this point General Stratemeyer, commanding the Army Air Forces, 
India-Burma Sector, intervened and upset the planned schedule of resupply 

58 Hunter tells, in his notes on the draft of this chapter, of his effort to prearrange through 
Merrill for the immediate supply of food and ammunition, and for the arrival of infantry to exploit 
the dazzling success that would be gained were the strip captured quickly in good condition. 

,9 (1) Stilwell Diary, 17 May 44. (2) The St i I well Papers, p. 296. (3) G-3 Per Rpt, Hq Chih 
Hui Pu, 24 May 44. NCAC Files, KCRC. (4) NCAC History, Vol. I, App. 2, The Siege of Myit- 
kyina, pp. 13-17. (Hereafter, this section of the NCAC History will be called Siege of Myitkyina.) 
(5) Hunter Notes. 



and reinforcement by ordering the W and X Troops of the 69th Light (Anti- 
aircraft) Regiment, a British unit, to be flown in. When these arrived on the 
18th, Merrill and Hunter were sorely disappointed to receive more antiaircraft 
instead of the infantry and supplies so badly needed if the town was to be taken 
quickly. There could be no quarrel with Stratemeyer's intent, for the precau- 
tion was a reasonable one, but the local commanders would have preferred the 
arrangements they themselves had made. 60 

Back at his headquarters Stilwell was exultant. Again and again he had 
been told that Myitkyina could not be taken, if taken could not be held, and if 
held was not worth holding. Now his transports were landing on Myitkyina 
airstrip, flying in the Chinese who would take the principal center of north 
Burma and go on to lift the blockade of China. In six months his forces had 
driven 500 miles into Burma and won engagements against seven Japanese 
regiments, among them, the victors of Singapore. The brilliant seizure of the 
Myitkyina airstrip was the height of Stilwell's career and the grand climax of 
the North Burma Campaign. 61 

The seizure of the principal airstrip at Myitkyina on 17 May was a stunning 
surprise to SEAC. The Prime Minister sent a radio to Admiral Mountbatten 
asking if he had expected the sudden blow at Myitkyina. In reply, Mountbatten 
pointed out that he had heard only incidentally that Stilwell planned to attack 
Myitkyina. Since this attack might eventually involve the use of considerable 
numbers of British troops, Mountbatten thought that he should have known 
beforehand of the decision to attack Myitkyina and planned to send Stilwell a 
personal letter explaining his position. 62 

Stilwell's coup cast a new light on the long-debated SEAC directive. As 
the Allied columns drew ever closer to Myitkyina, the British and American 
views on the North Burma Campaign had been approaching agreement. The 
Strategy and Policy Group of OPD on 3 May had noted that the speed of the 
American advance across the Pacific promised to outstrip any ground action 
that might be undertaken in China, which suggested to them that Stilwell 
should be told to concentrate on building up his air force in China to support 
the U.S. offensive in the Pacific. 63 The current position of the British Chiefs of 
Staff was an acceptance of Mountbatten's view that SEAC's directive should be 
to develop, maintain, broaden, and protect the air link to China with its current 
resources, in time to support projected operations against Formosa. 64 

The Joint Staff Planners of the JCS on 16 May suggested adopting the 
British position. That same day, the Army Service Forces, which had been 

60 (1) Ltr citec pT. 16| (2) Hunter Notes. 

61 The Stilwell Papers, p. 296. 

62 (1) Min, SAC's 104th Mtg, 20 May 44. SEAC War Diary. (2) This letter cannot be found 
among Stilwell's papers, and perhaps was not sent. 

65 Memo, Roberts for ACofS, OPD, 3 May 44, sub: Memo for Gen Somervell Concerning Plan 
of Action Within China. Case 367, OPD 381 Security, A47-30. 
64 JPS 406/3, 16 May 44, sub: Strategy in SEAC. 



energetic in support of the Ledo Road, argued that the Joint Planners were in 
effect suggesting an end to offensive action in north Burma at a time when the 
trace of the Ledo Road was virtually in Allied hands. Then came Stil well's bold 
stroke at Myitkyina. In response the planners decided to suggest that acceptance 
of the British position be contingent on securing the Mogaung-Myitkyma 
area/ 5 ' 

The result of these different points of view, powerfully affected by the blow 
at Myitkyina, was a compromise CCS directive to SEAC, issued 2 June 1944. 
Unlike the Quadrant directive, which had taken no stand on ground versus 
air operations in Burma, the new CCS directive gave first priority to building 
up the Hump operation in order to provide the maximum flow of supplies to 
the air force in China for support of Pacific operations. So far as was consistent 
with the primary objective, SEAC was to press advantages against the enemy 
by exerting the maximum ground and air effort, and in so doing be prepared to 
exploit the opening of the Ledo Road. 

Securing Myitkyina, conquering enough of north Burma to protect the 
Allied hold on that key point, and building pipelines to China were viewed as 
integral parts of building up air strength in China, By omitting ground opera- 
tions save as they did not conflict with the Hump build-up, the CCS tempo- 
rarily resolved the conflict over the location of a ground offensive. In this way 
the controversy over operations in north Burma as against those in Sumatra 
(Operation Culverin) was settled for the present, and perhaps for good. 66 

The Myitkyina airstrip was not long in proving its worth, and as the weeks 
went on more and more troops, food, ammunition, artillery, and construction 
equipment were flown in to carry on operations against the town of Myitkyina, 
to build the line of communications across north Burma, and to support the 
transports that made Myitkyina a base on the way to China. From May to 
October 1944, about 14,000 transport flights into Myitkyina were logged, 
carrying over 40,000 tons of cargo. 67 

The First Attempts To Take the Town 

Initial attempts to take the town of Myitkyina were hampered by poor 
intelligence and faulty organization. At Myitkyina the local intelligence 
agencies gained the mistaken impression that they were faced by relatively few 
Japanese^ and plans wete made accordingly. On 18 May the total of Japanese 
was set at 300; on 15 June aftet a period of uncertainty G-2 of the Myitkyina 

« Memo, Deputy Director, Plans and Opns, ASF, for ACofS, OPD. 16 May 
in SEAC; Memo, Roberts for CG, ASF, attn Deputy Director, Plans and Opns, 
sub: Strategy in SEAC; JCS 7 74/1, ABC 384 (Burma), 8-25-42, Sec 6, A4S " 
this paper on 26 May 1944 as JCS 774/1. 

&m Mm, CCS 161st Mtg, 2 Jun 44. (2) CM- OUT 46159, Marshall so Stilwell. 4 Jun 44. 

* T EAC Despatch, p. 115. 

sub: Strategy 
^19 May 44, 



Task Force set the number at 500, a gross underestimate. 68 There was much 
confusion in the first few days at Myitkyina. It was aggravated by the fact that 
the Chinese troops came from three separate divisions, the 30th, 14th, and 50th. 
The toll it took was heavy. 

On the afternoon of 17 May, after the airfield was cleared of Japanese, two 
battalions of the 150t h Regiment moved off on the first attempt to take 
Myitkyina. | (Map 16) Following the wrong road, they went toward Sitapur 
instead; on encountering Japanese rifle fire, they lost direction completely and 
engaged in fighting among themselves. Meanwhile, Colonel Hunter was calling 
up K and M Forces, directing them to move at once. Although the men had 
not eaten for several days, they abandoned all attempts at receiving the airdrop 
they so badly needed, took up their packs, and set out for Myitkyina. 69 

As one Galahad company trudged on to the airstrip, Merrill noted that 
hardly a man could walk normally, for fatigue, Naga sores, and skin diseases 
("jungle rot") were making themselves pitifully evident. One platoon, 
tormented by dysentery, had cut away the seats of their fatigues so as not to be 
unduly hampered in combat. His "men were pitiful but still a splendid sight," 
wrote Merrill after the war. 

On 18 May the rest of the 89th Regiment and a company of heavy mortars 
were flown in. They were followed next day by the 3/42d of the Chinese 14th 
Division. Stilwell arrived to watch operations. 

About noon of the 18th the 150th Regiment of the Chinese 50th Division 
was sent toward Myitkyina and, repeating the mishap of the day before, became 
confused, fought among themselves, and drove themselves right back out of 
the town. In these episodes passed the opportunity of swiftly overrunning 
Myitkyina. Galahad personnel were now fatigued and riddled with tropicr! 
disease, just as the Chindits were, and no fresh regiment had been provided 
from the United States for Myitkyina, on whose capture the JCS had placed 
such emphasis. In the strain of these days Merrill suffered another heart attack 
and had to be evacuated. An assault on 20 May carried the 150th Regiment to 
the railroad yards, but then was stopped by heavy Japanese fire. The 150th, 
which by this time had taken 671 casualties since the campaign opened, had to 
be withdrawn. 70 

Japanese Build-up at Myitkyina 

The Japanese commanders at Myitkyina had been given the mission of 
holding there until mid- August. They were soon winning the race to build up 

68 (1) Galahad, p. 103. (2) G-2 Roundup, Hq MTF, 15 Jun 44. AG 319.1, NCAC Files, 
KCRC. This document quotes Galahad Force officers to that effect, but the estimate was picked 
up by Headquarters, Chih Hui Pu, in their G-2 Daily Situation Report 33 of 30 June, and as late 
as 24 July Brig. Gen. Russell E. Randall, arriving at Myitkyina, was told there were only 500 
Japanese there. (3) Rpts 470, 477, 478, JICA, 6 Sep 44. JICA CBI Br, Calcutta, India. OCMH. 

69 (1) Galahad, p. 98. (2) Siege of Myitkyina, p. 11. 

70 (1) Siege of Myitkyina, pp. 18-19. (2) Stilwell Diary, 18 May 44. 



JAPANESE DEFENSIVE POSITIONS in the railroad yards at Myitkyina. 

enough strength to carry out their orders. As early as 1700 on 17 May Colonel 
Maruyama had brought the main force of the 1/ 148th, 56th Division, in across 
the Irrawaddy. Other Japanese hurried in at night through the gaps in the 
Allied position about the city, so many that by 31 May there were about 2,500 
Japanese in Myitkyina plus several hundred sick and wounded. From the 56th 
Division on the Salween came the commander of its infantry group, Maj. Gen. 
Genzu Mizukami, with his staff, the rest of the 1/I48th, and two pieces of 
artillery. Assuming command, General Mizukami ordered that Myitkyina hold 
for at least three months and the surrounding area for thirty days more to keep 
the Chinese and Americans from moving south into central Burma. Control of 
Japanese operations at Myitkyina was assumed by 33d Army in early June, and 
Tanaka was freed to concentrate on operations around Kamaing. 71 

Fortunately for the attackers, the Japanese grossly overestimated their 
strength, as they seem to have done throughout the North Burma Campaign. 
It was thought that the Myitkyina Task Force comprised 30,000 men. The 
Japanese knew they inflicted heavy casualties but thought that replacements 

71 (1) Maruyama Interrog. OCMH. (2) Tanaka, Japanese Comments, Sec. III. (3) Japanese 
Officers' Comments, p. 19. 



were coming in at a corresponding rate. 72 Many of the Japanese sick took their 
places in the fortifications that were quickly built and fought with the tradi- 
tional stubborn valor of their people. Equipped with plenty of machine guns 
and mortars, and four mountain guns, they were well dug in, ready to die for 
the Emperor. From beginning to end, at least 4,600 Japanese fought at 
Myitkyina; the peak strength at any one time may have been 3,500 men. Their 
morale was high and it was long before they stopped believing that relief 
would soon arrive. 73 Indeed, at one time a regimental combat team of the 
53d Division under a Colonel Asano was actually under way to relieve 
Myitkyina but the Allied advance on Mogaung forced its recall. 74 

So, this was no handful to be brushed aside, especially not by troops 
fatigued by long exertion and privation, whose fire support was initially a 
battery of pack howitzers plus what the fighter bombers could do. Perhaps 
luckily, these steadily worsening odds were unknown to the attackers. On 19 
May Merrill estimated there were two and one-half battalions of Japanese in 
Myitkyina plus more coming up from the south. Stilwell's G-2, Colonel 
Stilwell, was skeptical. Certainly on the 19th the organizational framework of 
two and one-half Japanese battalions was in Myitkyina but the numerical 
equivalent of such a force was not. 75 

Hoping for a quick seizure of the town, which would permit subsequent 
operations against the rear of the Japanese forces on the Salween front, Stilwell 
briefly considered asking that the British 36th Division be rushed in to take 
Myitkyina. Giving no reason in his diary, he decided against this move and 
instead resolved to order in some U.S. combat engineers from the Ledo Road. 
"I will probably have to use some of our engineer units to keep an American 
flavor in the fight," he told Marshall. 76 

Galahad took the suburban village of Charpate, north-northwest of 
Myitkyina, and the 88th Regiment extended its lines south to the railway 
tracks, which in this area run northwest before turning south to Mogaung. 
Unfortunately, K Force did not occupy a small height which dominated 
Charpate. M Force took the little settlement of Namkwi without opposition, 
in order to shield the Allies against any Japanese attempt to relieve Myitkyina 
by an attack from Mogaung. On 21 May the 3d Battalion of Galahad was 

72 Maruyama Interrog. OCMH. 

7 ^ (1) Galahad, pp. 104-05. (2) PW History. (3) Statement, Sgt Maj Hirose Yutaka, 10 
Aug 44. Folder 62J, 161-170. NCAC Files, KCRC. A code clerk, Yutaka was an excellent source 
of information on the Japanese side. (4) Maruyama Interrog; Tanaka Interrog. OCMH. (5) 
SEATIC Bull 247, pp. 18-19- MID Library. (6) A G-2, Myitkyina Task Force, memorandum, 
9 August 1944, estimated that 4,075 Japanese were killed at Myitkyina. These, plus the several 
hundred who escaped, come close to Tanaka's figure of 4,600. The garrison commander, Maru- 
yama, set his strength at 3,500, which can be reconciled with the above if Tanaka's figure is taken 
as the grand total. The G-2 estimate is in Wessels' File, OCMH. 

74 SEATIC Bull 244, p. 2. MID Library. 

75 Col Stilwell's Diary, 19 May 44. 

76 (1) Stilwell Diary, 22 May 44. (2) Rad CHC 1097, Stilwell to Marshall, 22 May 44. 
SNF 131. 



ordered by the newly created Headquarters, Myitkyina Task Force, to seize the 
auxiliary airstrip north of Myitkyina. Next day, Col. John E. McCammon was 
put in command. At his disposal were the 88th and 89th Regiments of the 30th 
Division, Gen. Hu Su commanding, an improvised division of the 150th and 
the first echelons of the 42d Regiment, under Maj. Gen. Pan Yu-kun of the 50th 
Division (who arrived 23 May), the survivors of Galahad, the Seagrave Hos- 
pital Unit, and the 42d Portable Surgical Hospital. 77 For the occasion, Stilwell, 
stretching several points, told McCammon to assume the rank and insignia of 
brigadier general and rushed off a recommendation for promotion to the War 
Department. McCammon's assignment was a difficult one, doubly so for a man 
just out of hospital. Keeping his stars in his pocket, and assisted by Hunter as 
executive and commander of Galahad, McCammon took up his task. 78 

McCammon's command was a little island of precariously held territory, 
which lay between Mizukami's garrison in Myitkyina and the Japanese forces 
to the south, north, and west. Across the Irrawaddy to the east was a small force 
of Chindits, Morris Force, with orders to block Japanese traffic from the east, 
a task of which it proved incapable. McCammon's forces were completely 
dependent on air supply, which in the early weeks kept them in about two 
days' reserve supplies. 

The main airstrip was tightly held by the 150th Regiment, whose lines 
extended toward the Irrawaddy. North of it and to the west were the perimeter 
defenses of two Galahad combat teams. The 88th and 89th Regiments were 
on either side of the tracks of the Myitkyina-Mogaung railway, and the 3d 
Battalion of Galahad was north of Myitkyina. All units were deployed into 
strongpoints organized for all-round defense. There was much patrol activity 
during the daylight hours to keep clear the trails over which oxcarts moved 
supplies from the airstrip. 

Gaps were numerous through which the Japanese could work their way 
into Myitkyina, and Mizukami set up a regular ferry service across the 
Irrawaddy by which supplies and individual replacements could be brought in. 
The terrain was excellent for defense. The roads lay high above the surround- 
ing rice paddies, and each was therefore an earthwall making a first-rate 
obstacle. Clumps of trees were all about, and there were plenty of houses on 
the outskirts of town which the Japanese used to advantage. The northern part 
of the Japanese perimeter was well shielded by a crescent-shaped depression 

77 (1) Galahad, p. 106. (2) Siege of Myitkyina, p. 22, and App. (3) Stone MS, pp. 187-89. 

78 ( 1) Probably McCammon was given this rank to help him command the Chinese. Merrill 
recalled suggesting to Stilwell that a Chinese officer be placed in command at Myitkyina. There is 
no evidence from which to reconstruct Stilwell's intentions, but McCammon's elevation marked 
the sixth time Stilwell had used American officers to command Chinese troops in Burma, McCam- 
mon's predecessors being Boatner, Col. Rothwell Brown, Col. Campbell Brown, Colonel Sliney, 
and Colonel Dupuy. See Merrill Notes. (2) Notes by Col McCammon on draft MS of this chapter. 
(3) Ltr, Dupuy to Ward, 12 Sep 52. Colonels Sliney and Dupuy were authorized to give 
commands to Chinese artillery in Stilwell's name. 



JAPANESE TRENCHES at the outskirts of Myitkyina. 

which the monsoon rains quickly turned into a swamp. In effect, the Allies 
and the Japanese were committed to position warfare. Though the operation 
was called a siege, the Japanese received a steady trickle of supplies and 
replacements until Mogaung fell. 

The problem facing the Allied command was to overcome the combination 
of machine guns and earthworks which had been so effective on the Western 
Front in Europe in 1914-18. They would have to do it without tanks, with a 
final maximum of fourteen artillery pieces, and with air support (the twelve 
P-40's of the 88th Fighter Squadron, ultimately based on the airstrip itself) 
which, though devoted and skillful, did not have the weight to drive the 
Japanese from their positions. To make matters completely uncomfortable and 
further complicated, after mid-May the rains grew steadily worse, which made 
air supply very difficult. It was Cassino on a shoestring. 

The first phase of the operation was an attempt to occupy the town by 
moving in. It was dominated by the belief that Myitkyina was lightly held. 
The 3d Battalion, Galahad, attacked southward on 21 May, but while still 
far from the northern airstrip was forced to dig in at the Mankrin-Radhapur 
road junction. That night it was attacked from the rear, or Mogaung, side and 
the following day fell back to Charpate. The night of the 23d the Japanese 



struck again, and the tired, fever-ridden men of the 3d Battalion repelled the 
attack with difficulty. Another attack came the next morning, and the 3d was 
pushed back. These were ominous signs of the shifting balance of strength. 
There was yet another in Galahad's evacuation rate, now running between 
75 and 100 a day. At this time, from 15 to 30 men a day began to report sick 
with symptoms of the lethal mite or scrub typhus, and about 80 percent of 
Galahad suffered from dysentery in various forms. 79 

Colonel McCammon's first attempt at a full-dress assault was made on 24 
May. The 88th and 89th Regiments were ordered to jump off at 0700 on the 
25th and drive through Myitkyina to the riverbank. When the day ended, the 
88th Regiment had succeeded in straightening its lines; the record does not 
even mention its sister regiment. McCammon, who was in the early stage of 
pleurisy, was depressed by the failure, and Stilwell flew in to check the situation. 
While he was there, Hunter gave him a letter stating that Galahad was being 
unfairly treated, and that under Boatner's influence Headquarters, NCAC, was 
discriminating against Galahad in favor of the Chinese. 80 After a brief return 
to his headquarters near Shaduzup Stilwell went again to Myitkyina and made 
a quick decision, relieving McCammon and replacing him with General 
Boatner on 30 May. 

While Stilwell was making these command changes, the earlier optimism 
was replaced by a brief period of extreme alarm, caused by the rapid disintegra- 
tion of Galahad and the Chindits' evacuation of the block they had placed 
across the railway near Hopin. Only twelve men were left in the 2d Battalion of 
Galahad, while the Chinese 150th Regiment was down to 600 men. 81 

American reinforcements of any men who could hold a rifle were rushed in 
from every possible source. Between 26 May and 1 June, two engineer bat- 
talions, a group of replacements for Galahad (optimistically assembled as a 
battalion), and evacuees of Galahad itself were flown in. With the situation 
thus steadied and with Stilwell and Chih Hui Pu believing that Myitkyina was 
held by only a few hundred Japanese, a plan was made for a co-ordinated 
attack by two Chinese regiments plus the survivors of Galahad and the 209th 
U.S. Engineers. In this operation, which was attempted on 31 May, the 42d 
Regiment reached the Waingmaw ferry road. Built up twelve feet above the 
neighboring paddy fields, the road gave the Japanese a magnificent defensive 
position, which they exploited cleverly. The Chinese recoiled from this natural 
fortification but were able to beat off a Japanese counterattack. The 150th Regi- 
ment reached the riverbank and drew up in an arc about a sawmill in which 
the Japanese had a strongpoint. The 88th and 89th Regiments were not in- 

79 (1) Galahad, p. 106. (2) Siege of Myitkyina, p. 25. (3) Stone MS, p. 190. 

80 (1) Hunter Notes. (2) Boatner Notes. (3) Stilwell's only diary reference of this period to 
Hunter is for 26 May: "Hunter, Osborne, McGee, Beach— all fine soldiers. Guts, calm, confidence. 
They ooze it." (4) Ltrs, Boatner to Stilwell, Jun 44. SNF 207. The letters praise Hunter highly. 

81 Stilwell Diary, 30 May 44. 



volved in this attack. The American attempt netted little, and it was apparent 
that Myitkyina would not fall at once. 82 

The night of 31 May the Japanese attacked the 209th Engineer Battalion in 
position near Radhapur, north of Myitkyina. After the war Colonel Maruyama 
stated that on 30 May he tried to break out with three battalions. General 
Tanaka recalled that such an attempt had been planned but canceled. Whatever 
the mission of these particular Japanese, there was heavy, close-in fighting on 
the Myitkyina side of the engineers' perimeter. 83 When the day came, the 
Chinese tried again to take Myitkyina but counted their gains in yards. The 
236th U.S. Engineer Battalion, which on twenty-four hours' notice had been 
taken from its road building and from the operation of gravel and ice plants 
and rushed to Myitkyina, was sent to retake Namkwi, west of the auxiliary 

The motive behind the 236th's attack of 1 June was to contain the Japanese 
in the Namkwi area and introduce the battalion to combat under relatively 
easy conditions. One company of the 236th did succeed in entering Namkwi 
but instead of promptly consolidating to meet the inevitable Japanese counter- 
attack fell out for a break. The Japanese counterattacked and drove the unwary 
engineers right back out of the village. 84 

Next day, 2 June, the Chinese commenced formal siege operations in the 
eighteenth century manner, tunneling toward the Japanese; in the virtual 
absence of artillery there was nothing else to do. General Boatner was still 
optimistic on 2 June, but after that time the Allied command again feared the 
Japanese might yet turn the tables on their attackers. Reinforcing this 
impression was the steadily graver aspect of local command problems. 85 

Command Problems at Myitkyina 

The steady deterioration in the physical condition and morale of Galahad, 
the cumulative effect of the fighting on the strength of Chinese units, the poor 
combat performance of the engineers and the Galahad replacements, the bad 
supply position, and the increasing aggressiveness of the Japanese brought 
about a period in which the optimism of a few days before yielded to the fear 
that the Japanese might overrun the airstrip and win a major victory. On 20 
May the Allied forces had three days' rations on hand, very little mortar am- 
munition, and only 350 rounds of 75-mm. shells. Six days later nine disabled 
transports on the airfield mutely added another reason for the supply pinch. 
At the end of May the quartermaster had no U.S. rations, while 40 percent of 
the rice and grain rations for the thousands of Burmese, Chinese, and refugees 
from Myitkyina was spoiled by dampness. The effect of the supply situation 

82 (1) FO 12, Hq Camelot Task Force, 30 May 44. NCAC Files, KCRC. (2) Siege of 
Myitkyina, pp. 30-31. 

83 Maruyama Interrog; Tanaka Interrog. OCMH. 

84 (1) Hunter Notes. (2) Siege of Myitkyina, p. 31. 

" (1) Siege of Myitkyina, pp. 28, 31-32. (2) Stilwell Diary, 2 Jun 44. 



was underscored when the force surgeon asked for some variety in the food 
because the American troops were vomiting all but the breakfast part of the 
K ration. Attempts were made to control ammunition expenditure, since the 
Chinese often seemed extremely wasteful in the use of it. Captured Japanese 
weapons were being turned against their former owners. By 1 June the food 
supply was set at one day for Americans and two for Chinese. There were now 
thirteen damaged transports on the airstrip. 86 

The Galahad situation became obviously critical around 19 or 20 May. 
While Stilwell had been ordering the 3d Indian Division kept in Burma over 
the protests of its commanders that their men were exhausted, the Galahad 
evacuation rate had been steadily expanding up to 134 in one day. As set by 
Hunter, who had received no directives and had to exercise his own best 
judgment, Galahad policy required for evacuation a fever of 102 degrees for 
three consecutive days, and approval of each case by a board of doctors. This 
was a delicate situation. 87 

Boatner was under a certain handicap in his relations with the American 
troops. He knew of Hunter's letter of 27 May, with its charges against him, 
charges he felt would not have been made if it had been generally known that 
he had been in the United States during most of the campaign. He believed that 
Galahad enlisted personnel fully endorsed the charges made. After reflection, 
he decided that his contacts with Galahad and Hunter should be through 
his own chief of staff, Col. John P. Willey. A further argument was that Willey 
spoke no Chinese, while Boatner knew the language and considered he would 
be fully occupied in trying to exercise command over the Chinese, drawn as 
they were from three divisions. 88 

With the growing realization that Myitkyina would not fall quickly, that 
the Japanese might take the initiative, Stilwell ordered staff officers in the rear 
area to send any Galahad personnel fit for duty back to the fight. This order 
was in sharp contrast to the men's expectation that after reaching Myitkyina 
they would have a long period of recuperation. 

General Boatner gave four reasons why this unpleasant and difficult decision 
was made: 

1. Galahad was the only U.S. combat unit in the theater available for the 
assault on Myitkyina. 

2. The Chinese 88th and 150th Regiments that marched over the Kumon 
Range with Galahad had had few evacuations for sickness or fatigue in spite 
of their heavy casualties. 

3. Since early May Stilwell had been resisting heavy pressure to evacuate 3d 
Indian Division. 

86 Rpts, Maj Milligan Bethel, G-4, to Lt Col Joseph A. McNerney, Supply and Evacuation Off, 
G-4, 20 May-1 Jun 44, sub: Supply Rpts. NCAC Files, KCRC. 

87 (1) Memo, Boatner for Col Stanley F. Griswold, 25 Jun 44. Folder, Misc Corresp MTF, 
NCAC Files, KCRC. (2) Galahad Investigation. (3) For Stilwell's authority over the 3d Indian 
Division, seepage 221, above. 

88 Boatner Notes. 



4. The Japanese lines were only 1,500 yards from the airstrip which was the 
only base and source of supply. 

Stilwell, said Boatner, thought it necessary not to "let the impression be 
created that we were withholding U.S. troops from combat in a sector where 
as an Allied commander Stilwell was keeping British and Chinese troops in 
combat." 89 

The exhaustion of Galahad personnel was appearing in combat. When 
the Japanese counterattacked at Charpate, 27 May, Colonel McGee's men were 
so tired that they kept falling asleep during the engagement. McGee himself 
fainted three times during the battle but with iron determination commanded 
his battalion from the aid station. McGee asked that his unit be relieved. 90 

Meanwhile, in the rear areas efforts were being made to round up Galahad 
evacuees who were fit for combat. Extremely heavy moral pressure, just short 
of outright orders, was placed on medical officers to return to duty or keep in 
the line every American who could pull a trigger. In one group of 200 men 
sent to Myitkyina many were not fit for duty, and ten were immediately 
re-evacuated. This incident, plus the fact that still more Galahad survivors 
were actually ill with malaria and dysentery, seemed to mark the exhaustion 
of the moral as well as the physical reserves of Galahad. Summarizing this 
unhappy episode, the inspector general wrote that the plans and assumptions 
of the War Department plus unauthorized statements reached the enlisted men 
and junior officers of Galahad as promises of what was in store. These never 
materialized and, coupled with "the physical deterioration of the unit, after 
months of arduous jungle combat and culminating in a rapidly growing feeling 
that hospitalization procedures were not being carried out, resulted about 
June 1st 1944 in almost complete breakdown of morale in the major portion 
of the unit." 91 However, it should be noted that six Distinguished Service 
Crosses went to Galahad veterans during the siege and Galahad itself was 
awarded the Distinguished Unit Citation. 

By 4 June, Galahad casualties were: 

Casualty Number 

Deaths 93 

Nonbattle deaths 30 

Wounded 293 

Missing 8 

Amoebic dysentery 503 

Scrub typhus 149 

Malaria 296 

Psychoneurosis 72 

Miscellaneous 950 

89 (1) Memo cited |n. 87 ( 1 ) J ( 2 ) Galahad Investigation. 

90 Galahad, p. 108. 

91 Galahad Investigation, p. 3. 



Figures for wounded in action are deceptively low because minor cases were 
treated in the unit. This is also true of malaria, which in many cases was not 
approved as a cause of evacuation. Other figures represent men actually evacu- 
ated. The average loss of weight per man was estimated at thirty-five pounds. 
It was observed that chlorinating water with the means available for individuals 
or small groups offered little protection against amoebiasis. The circumstances 
were different with respect to malaria; the men could have taken atebrin. The 
general appearance of malaria at Myitkyina meant a breakdown in morale. 92 
The engineers rushed to Myitkyina from their rear-area duties were no 
substitute for the trained, organized, and experienced Galahad combat teams. 
Their performance and that of the Galahad replacements who began to 
arrive was what might be expected from raw troops not acquainted with each 
other or their leaders. Following an incident in which a company of the 209th 
Engineers broke and ran after being ambushed by a small force of Japanese, the 
209th's companies were sandwiched between the Galahad units. Stilwell told 
Marshall that the engineers ran on several occasions, "incidentally abandoning 
wounded." On 8 June, the 209th and 236th were grouped as a provisional 
regiment and, once shaken down, fought with great credit to themselves and 
their corps. 93 

The Galahad "replacements" had greater potentialities than the engineers, 
for the latter had not seen a rifle since their basic training days and had simply 
been taken from their bulldozers and power generators to fight as infantry 
combat teams. As of 28 May about 2,600 replacements were in India. Stilwell 
ordered them rushed into action, for he feared that the Chindits' withdrawal 
from the block they had put across the railway near Hopin would bring down 
the Japanese in force. The War Department's opinion in October 1943 had 
been that refilling Galahad by individual replacements would be impracti- 
cable. Instead, the War Department proposed to form new units in the United 
States and ship them out intact. When they arrived in the theater, Galahad 
veterans were to be assigned to key posts in the new units. 

At Fort George G. Meade, Md., the replacements were set up (on the "four 
platoons to a company" basis) as an infantry regiment less headquarters and 
headquarters staff. The War Department intention behind this preliminary 
step was apparently never made clear to the lower administrative echelons, for 
the battalions and companies were never activated nor allowed to function as 

From the text of the call for volunteers that brought many men into the 
body of replacements, many believed their destination to be Burma. Because of 
the imminence of combat the advisability of letting the ad hoc companies 

92 Stone MS, pp. 197-98. 

93 (1) Rad CHC 1216, Stilwell to Marshall, 22 Jun 44. SNF 131. (2) Ltrs, Boatner to Stilwell, 
3, 5, 7, 12, 14, 15 Jun 44. Folder, Misc Corresp MTF, NCAC Files, KCRC. (3) Leslie Anders, The 
Engineers at Myitkyina, MS. Engineer Hist Sec. 



formed for shipping purposes function as bona fide infantry companies was 
several times called to the attention of proper authority by individual officers at 
Fort Meade, at the port of embarkation, and even on the transport, but the 
several headquarters would not concur. For its part, theater headquarters in 
May 1944 was unaware of these points. 94 

Of the 2,600 replacements, 400 formed two quartermaster pack trains. The 
rest were mostly infantrymen, with some artillerymen. There was a generous 
seasoning of experienced commissioned and noncommissioned officers who 
had volunteered from the permanent cadre of the several replacement centers. 

On arrival at Bombay on 25 May, the men were speeded by train to 
Ramgarh. After an average stay of about one week (for some men were at once 
flown to Myitkyina as individual replacements), the men were hastily formed 
into two battalions, flown to Myitkyina, and placed in combat as the "New 
Galahad." Unfortunately, filling the Table of Organization spaces of a bat- 
talion with men does not produce a fighting team until the men have had a 
period of training together. The commanders at Myitkyina took what corrective 
measures they could. Three battalions were formed of the survivors of the old 
Galahad and the newcomers. Boatner listed them as the 1st Galahad (old), 
2d Galahad (new), 3d Galahad (new). Officers and men were freely 
exchanged among the battalions to spread the hard-won experience of the 
Galahad veterans. The resulting strengths were: 1st Battalion, about 300 
men; 2d, about 950; 3d, about 950. 95 

A training program was also set up. Lessons were given in the intervals 
between combat. The men were also sent to the Namkwi area to shoot at live 
targets:". . . the Japanese disposition there remained defensive and the replace- 
ments could be disengaged without being pursued." 96 The shock of an intro- 
duction to combat under the conditions then prevailing at Myitkyina produced 
about fifty psychopathic cases among the replacements, and some of the officers 
were unfit. 97 

As a result of these circumstances, Boatner told Stilwell on 15 June: 

Reports continue to indicate the complete disorganization and fear in U.S. units. They 
are in many cases simply terrified of the Japs. We can expect time, experience, and casualties 
to reduce their strength but make those that remain better soldiers. From Galahad's rear 
[echelon] in Dinjan I hear that 250 of Old Galahad men are being equipped and will be 
flown back today. They will be of tremendous help. Rumor has it that they were moving 
around the countryside and many AWOL's. Col. Osborne just saw me and spoke most 
earnestly about how he felt these men are malingering and wants to go back to get a few 
more officers and many men back here. If Hunter OK's I will send him back. On the face 

94 (1) Rad WAR 3495, Marshall to Stilwell, 2 Oct 43. Item 1056, Bk 4, JWS Personal File. 
(2) Statement, Maj George L. Converse, 10 Jan 49. Major Converse was a member of the replace- 
ment shipment. (3) Rad CRA 4407, Sultan to Stilwell, 27 May 44. Item 185, Bk 6A, JWS 
Personal File. 

95 Ltrs, Boatner to Stilwell, 7 Jun, 10 Jun, 44. SNF 207. These letters tell of the reorganizing 
process of Galahad. 

96 (1) Siege of Myitkyina, pp. 31-35. (2) Statement by Converse cited n. 94(2). 

97 Rad CHC 1216 cited n. 93 ( 1 ) . 



of it it might appear that we have plenty here on the field for protection. Such is not actually 
the case— we in fact have only a prayer. 98 

Great though the endurance and devotion of the Chinese were, the steady 
drain of casualties was reducing their strength to dangerously low levels. This 
situation was called to General Boatner's attention on 7 June by the com- 
manders of the 30th and 50th Divisions, who pointed out that the 150th and 
89th Regiments mustered between them only 1,000 men. The 88th and 42d 
could each find only one battalion for an attack. The accuracy of their state- 
ments was confirmed by an American liaison officer, who reported that the 
regiment to which he was attached was down to 481 officers and men. Boatner 
relayed this information to Stilwell, adding that the Chinese troops were taking 
casualties (121 on a comparatively quiet day) while he was trying to make the 
raw Americans battleworthy. Obviously, the two engineer battalions and the 
Galahad replacements represented the margin of numerical superiority over 
the Japanese garrison of Myitkyina. But because of the fixed belief that there 
were only 500 Japanese in Myitkyina the full import of this was not realized; 
nonetheless, there was disquietude." 

Across the Irrawaddy from Myitkyina was a group of Chindits, Morris 
Force, that had been attached to Boatner's headquarters. Its experiences had 
paralleled those of Galahad and its companions of the 3d Indian Division, 
and like them it was disintegrating from the cumulative strains of the cam- 
paign. Commanded by Brigadier G. R. Morris, these 1,300 men were originally 
ordered to cut the Japanese line of communications via Bhamo to Myitkyina. 

On 25 May the orders were changed to clearing the Japanese from the east 
bank of the Irrawaddy opposite Myitkyina. The Japanese garrison in Waingmaw 
was a reinforced company, the terrain was flooded, and Morris's men were 
tired. Boatner welcomed them enthusiastically: "... really believe you are just 
in time for the kill." Morris did not take Waingmaw, and Boatner found it 
increasingly difficult to keep in touch with his headquarters. 

The tone of Boatner's communications grew ever stronger as he told Morris 
that Mountbatten had ordered Myitkyina taken at all costs, which meant 
having casualties in the effort. Eleven days later, on 14 June, Boatner told 
Morris that the situation was "precarious," that Morris was to attack any way 
he chose so long as it was promptly. Morris's replies then began to come 
through. He explained that except for a few avenues of approach, well covered 
by Japanese machine guns, the country was flooded chest high. His men were 
wasting away at the rate of one third of a platoon a day, were exhausted, and 
were falling asleep under Japanese fire. 

98 Ltr, Boatner to Stilwell, 15 Jun 44. SNF 207. 

99 Ltr, Gens Hu Su (30th Div) and Yu Pan-kum (50th Div) to CG, MTF, 7 Jun 44, sub: 
Recommendations for Opns (Urgent); Ltr, Gen Hu to CG, MTF, 7 Jun 44, sub: Rpt; Ltr, Boatner 
to Stilwell, 8 Jun 44; Ltr, Maj Edward H. S. Wilkie, Jr., to CG, MTF, 8 Jun 44. Folder, Misc 
Corresp MTF, NCAC Files, KCRC. 



By 14 July Morris had only three platoons left, and Lentaigne asked for his 
evacuation. Chih Hui Pu objected, saying Morris could still make an offensive 
effort. A week later Lentaigne asked ironically if Morris Force could be evacu- 
ated when it was down to 25 officers and other ranks from the 1,301 it had on 
19 May. Stilwell approved. 100 

The Attacks of Mid-June 1944 

Coming to a head almost simultaneously, these difficulties and dangers 
made General Boatner's position unenviable. Airborne assault and LRPG war- 
fare have the vices of their virtues, among them that the attacker cannot fall 
back and reorganize for a second try. Victory is the only solution to his prob- 
lems. There were days in which a banzai charge by General Mizukami's 
garrison or a determined push by the 33d Division (which had once been ordered 
to send a regiment and lift the siege) would in all probability have swept right 
over the airstrip. Even fairly accurate artillery practice by Mizukami's four 
75-mm. pieces, at about 2,500 yards' range, could have destroyed transports at 
a rate to make supply prohibitively expensive, while Boatner's headquarters 
itself was only a mile from the Japanese lines. 

Boatner made one more attack on 3 June with the 42d and 150th Regiments 
plus the l/89th. The two regiments took 320 casualties, but putting a good 
face on matters, Boatner reported to Stilwell that if he had had air support 
about noon he could have taken the town. Because of the Chinese casualties, 
the fall of his 75-mm. ammunition stock to 600 rounds, and his .303 and belted 
.30-caliber to one day's supply, Boatner reported that he would hold back until 
his supplies were built up. He reaffirmed his position a few days later, saying 
that the lack of an immediate Japanese threat, the need to cut Chinese casual- 
ties and to train U.S. troops inclined him to wait a few days more. He thought 
that the Japanese garrison was in bad condition. 101 

100 ( 1) Ltr, Hq Chih Hui Pu to CofS, 3d Ind Div, 16 Jan 44, sub: Addition to Dir for CG, 3d 
Ind Div; Ltr, Hq Chih Hui Pu to CG, 3d Ind Div, 25 May 44, sub: Dir to CG, 3d Ind Div. 
Folder, Memos and Opnal Instr, CAI NCAC, 3d Ind Div, Jan-May 44, NCAC Files, KCRC. (2) 
Ltr, Boatner to CO, Morris Force, 1 Jun 44, sub: Orders; Ltr, Boatner to Morris, 3 Jun 44, sub: 
Orders; Ltr, Boatner to CO, Morris Force, 7 Jun 44, sub: Orders; Rad M329, Boatner to Morris, 
14 Jun 44. Folder, Orders MTF, 3d Ind Div, Jun 44, NCAC Files, KCRC. (3) Ltr, Boatner to 
Morris, 2 Jun 44; Rad M333, Boatner to Morris, 14 Jun 44; Rad, Boatner to Morris, 14 Jun 44; 
Rad, Morris to Boatner, 14 Jun 44; Rad 49, Col to Amity, 3d Ind Div, and KKO, 14 Jun 44; Rad, 
Wallace to Boatner, 14 Jun 44. Folder, Rads Wallace-Boatner, Morris Task Force, Jun 44, NCAC 
Files, KCRC. (4) Ltr, Boatner to Stilwell, 6 Jun 44. Folder, Misc Corresp MTF, NCAC Files, 
KCRC. (5) Ltr, Morris to GO Commanding, Myitkyina, and 3d Ind Div, 2 Jul 44; Ltr 2/2/G, 
Lentaigne to Advance Hq CAI, 14 Jul 44. NCAC Files, KCRC. (6) Ltr, Hq Chih Hui Pu to CG, 
3d Ind Div, 20 Jul 44, sub: Relief of Morris Force by Haswell Force; Ltr 7/4/G, Lentaigne to 
Advance Hq CAI, 20 Jul 44; Ltr, Hq CAI to CG, 3d Ind Div, 21 Jul 44, sub: Status of Comd, 
Morris Force and 26 Column. Folder, Radios— 3d Ind Div, NCAC Files, KCRC. (7) Statement, 
Sgt Suyeyoshi Tokuda, Myitkyina PW Rpts. PW History. 

101 Ltrs, Boatner to Stilwell, 3, 4, 5 Jun 44. Folder, Misc Corresp MTF, NCAC Files, KCRC. 



During the next few days the Allies pecked away at the Japanese lines, 
while the staff prepared orders for an attack on 10 June. This operation called 
for the 150th Regiment to take Japanese positions in the railroad yard running 
due east from the engine shed; the 42 d Regiment would go to the Irrawaddy; 
and the 89th Regiment, which had a zone twice that of the 42d and 150th 
together, would attack directly toward the river. Chinese units, lying west and 
south of Myitkyina, were to move northeast into the town, then wheel east and 
drive to the river. The Americans to the north would attack southwards. On 
the south side of the city was the bulk of the infantry, but the mass of the 
artillery supported the 30th Division (88th and 89th Regiments). Artillery 
was arriving by air now. Two batteries plus one platoon of 75-mm. howitzers; 
two 105-mm., and two 155-mm. howitzers, were ultimately present. All except 
two pieces with Galahad were kept under headquarters control. During the 
siege they fired 600 tons of ammunition, very rarely with massed fire. 102 

The attack jumped off as scheduled. Flying over the battlefield on 14 June, 
Boatner reported to Stilwell that he did not see much "effort" being made, 
though he found the scene a beehive of industry compared with the last week. 
On the ground matters seemed lively enough at the infantry's level. On 13 June 
the Japanese hit a platoon of K Company, New Galahad, so hard that the 
company broke and re-formed on the L Company line. The portion of the 
Japanese thrust that hit the perimeter next to the river made most of the men 
"take off," but two stayed in place and repelled the Japanese with an automatic 
rifle and a machine gun. To the west of this little break the Japanese worked 
their way in close but were stopped by grenades and small arms fire. 103 

In the course of the action between 13 and 16 June, a number of men in the 
small U.S. contingent distinguished themselves by their extraordinary bravery. 
Lt. Col. William H. Combs, who had been liaison officer with the Chinese 
150th Regiment, died of wounds received while attempting to warn units of 
the 209th Engineers against a Japanese ambush. Pvt. Howard T. Smith took 
command of his platoon when the platoon leader was killed and the attack 
stalled by a Japanese pillbox. Smith assaulted the pillbox singlehanded and 
silenced it with hand grenades. Pfc. Willard J. D. Lilly destroyed an enemy 
machine gun firing into his own machine gun position. While under fire 1st Lt. 
Melvin D. Blair rescued wounded and silenced an enemy machine gun. Smith, 
Lilly, and Blair were veterans of Galahad. When his company was pinned 
down by machine gun fire, T/Sgt. Richard E. Roe of the New Galahad 
crawled forward and at the cost of his own life grenaded a Japanese machine 
gun that was causing heavy casualties. Fatally wounded, he gave his life to save 

102 FO 14, MTF; Memo, Col Laughlin, 12 Aug 44, sub: Rpt on Artillery. NCAC Files, KCRC. 

105 (1) Ltr, Boatner to Stilwell, 14 Jun 44, SNF 207. (2) Rpt, Hunter, 13 Jun 44. Folder, Misc 
Corresp MTF, NCAC Files, KCRC. (3) For his valor on this occasion, Staff Sgt. Alvin O. Miller 
of the 209th Engineers received the Distinguished Service Cross. GO 99, Hq USAF CBI, 17 
Aug 44. 

ARTILLERY IN ACTION AT MYITKYINA. Afoiv, the Chinese crew of a 
10}-mm. howitzer, and below, a 75-mm. pack howitzer's GALAHAD crew firing on 
Japanese positions. 



his friends. Sgt. Fred N. Coleman of the 236th Engineers threw himself on a 
Japanese grenade, and saved two comrades. 104 

Despite these efforts and setbacks the impression persisted that Myitkyina 
was lightly held, with the Myitkyina Task Force G-2 Roundup circulating an 
estimate by "Galahad officers" that the besiegers faced only 500 Japanese, the 
remnants of two battalions. 105 

As a result of the attack, the 3d Battalion, Galahad, by 17 June, had cut 
the Maingna ferry road and reached the Irrawaddy north of Myitkyina. The 
150th Regiment in the same period finally took the sawmill and gained 200 
yards, using flame throwers. The 88th Regiment gained 100 yards from posi- 
tions in the old rifle range area of the Burma Frontier Force barracks. The 
principal gain for the Americans was the capture of the Myitkyina-Mogaung- 
Sumprabum road junction. The gains were not in proportion to the effort 
expended, and Stilwell ordered the end of all infantry attacks. Boatner replied 
that he would stop attacking Japanese positions until "... our troops are 
steadied and a favorable opportunity presents itself." 106 

There was reason for the troops to need steadying. A and B companies, 
209th Engineers, were cut off from their main body by infiltrating Japanese. 
Trying to close in on them, Company C and Headquarters and Service Com- 
panies were in turn halted by Japanese. The condition of A and B Companies 
became critical during 14 June, for they had only one meal with them. Two of 
their men managed to work their way back to the block on the Sumprabum 
Road with news of their plight, but enemy small arms fire prevented airdrops. 
The isolated companies finally made their way back in small groups to the rest 
of the battalion over 15 and 16 June. The 3d Battalion of Galahad reported 
trouble in effecting reorganization and enforcing orders. 107 

The Americans were not alone in their problems. Two companies of the 
Chinese 2/42d which had made a small penetration into the Japanese lines on 
14 June were wiped out by counterattack that night. These setbacks empha- 
sized the nature of the Myitkyina fighting. The Allies held a ring of battalion 
and regimental strongpoints enclosing a similar Japanese system. Though the 
Allied strongpoints were close enough for the troops in one to sortie to the aid 
of another should that be needed, they were not so close that interlocking fire 
could be put down to close the gaps. Consequently, there was plenty of room 
for maneuver and ambush, and the inexperienced engineers and New Galahad 
troops often suffered at the hands of General Mizukami's veterans. 108 

104 These men, among others, received the Distinguished Service Cross. For the leadership, 
professional skill, and valor he repeatedly displayed during the early days of the operation, Colonel 
Combs was posthumously awarded an Oak Leaf Cluster to the DSC. GO's 137, 78, 48, 131, Hq 
USAF CBI, 20 Oct, 20 Ju l, 21 Dec, 1 2 Oct 44. 

105 G-2 Roundup cite d n. bH I 2 1 1 

106 Ltrs, Boatner to Stifwell, l6 jun 44; G-3 Per Rpts, Hq Chih Hui Pu, 13, 20 Jun 44. NCAC 
Files, KCRC. 

107 Siege of Myitkyina, pp. 46-51. 

108 Siege of Myitkyina, p. 48. 



On 18 June Headquarters, Myitkyina Task Force, directed that, for the time 
being, tunneling would be used to close with the enemy. Patrol skirmishes 
continued, and energetic regimental and battalion commanders were occasion- 
ally able to carve out small gains, always against strong Japanese resistance. In 
one of the patrol clashes of this period, Pfc. George C. Presterly of the Engi- 
neers won the Distinguished Service Cross by a lone assault on a Japanese 
strongpoint. Moving out ahead of his patrol, he continued firing and advancing, 
drawing all the Japanese fire from his patrol. Even after being mortally 
wounded, Presterly continued his charge. 109 

On 18 June Stilwell made another quick trip to Myitkyina. Boatner had 
reported that "U.S. troops are shaky," possibly referring to the incident involv- 
ing the 3d Battalion of Galahad. Stilwell visited the lines at or opposite 
Charpate, Sitapur, and Mankrin. After discussing the situation with Colonel 
Hunter, now commanding the American forces under Boatner, Stilwell con- 
cluded that the "men looked good," that the picture Boatner had painted was 
"not nearly that bad." Chih Hui Pu G-2 on 30 June was still reporting only 
500 Japanese in Myitkyina and undoubtedly thought so a few days before. 
Stilwell was also of the opinion that Boatner should have spent more of his 
time with the troops. Adding these factors together, Stilwell ordered Brig. Gen. 
Theodore F. Wessels to fly from SEAC headquarters to Myitkyina. 110 

Changes in Command 

After General Wessels arrived, he spent about a week looking over the 
situation while Stilwell debated Boatner's relief. On 25 June Boatner had a 
severe recurrence of the malaria that had troubled him earlier in the campaign, 
and this made inevitable a change in command. 111 After a brief stay in the 
hospital, Boatner returned to duty as Commanding General, Northern Combat 
Area Command. 

Boatner's successor, General Wessels, had been on the staff of the Infantry 
School at Fort Benning, Ga., when he was sent to CBI as a part of Stilwell's 
plans for training the Chinese Army. When it finally became apparent that the 
Chinese did not accept these plans to an extent that would occupy all of the 
2,213 officers and men sent to CBI to train the Second Thirty Divisions in east 
China, Wessels was sent to SEAC headquarters. 

The billet Wessels took over on 26 June was an uncomfortable one, but the 
situation began to improve the day after he assumed command. Mogaung fell, 
and the Chinese forces from the Mogaung valley began moving up the railroad 
to connect with Wessels' forces. This was a great gift of fortune. It removed the 

109 GO 1 1 1, Hq USF IBT, 5 Jun 45, par. I. 

110 (1) Stilwell Diary, 13, 15, 17, 24 Jun 44. (2) The Stilwell Papers, pp. 304-06. (3) Daily 
Sitrep 33, Chih Hui Pu, 30 Jun. NCAC Files, KCRC. 

111 The Stilwell Papers, p. 306. 



GENERAL STILWELL AND COLONEL HUNTER talk to the leader of a patrol 
which has returned from Myitkyina. 

recurrent menace of a Japanese drive from Mogaung, guaranteed reinforcements 
and the opening of a ground line of communications, and meant that Wessels' 
men, instead of being an island in a hostile sea open to attack from 360°, could 
concentrate their attention on the Japanese to their front. Moreover, General 
Mizukami and his men lost one of the two bases (the other was Bhamo) from 
which supplies had trickled in to the Japanese. Wessels made a point of visiting 
each unit, talking with the men and trying to instill confidence and raise 
morale. 112 Under orders from Stilwell, Hunter on 29 June was placed in 
command of all U.S. troops at Myitkyina. 

Stilwell left Wessels a problem resulting from one of Stilwell's rare inter- 
ventions in the siege. After a conference with Hunter he personally ordered 
that a Chinese battalion of at least 400 men be sent to penetrate through the 
Japanese positions to Sitapur roughly on a southeast azimuth, cutting off the 
Japanese to the north from Myitkyina proper. The 1st Battalion, 42d Regiment, 
of about 250 men, made a very considerable advance over 28 and 29 June. It 
drove deep into the Japanese defense system, leading Stilwell to hope this was 

112 (1) Interv with Wessels, 19 Jul 48. OCMH. (2) History of India-Burma Theater, 1944- 
1945, I, 55. OCMH. 



the turning point; on receiving Japanese fire, it halted and dug in. Air supply 
was necessary. Since Stilwell had given Wessels personal orders to support the 
battalion, Company F of the 2d Battalion, New Galahad, was ordered to 
join the l/42d. 

F Company, unaware it had lost its way and under an inexperienced 
commander, proceeded with a small point almost directly ahead of the march- 
ing column. The company commander at the head of the point met a small 
group of Orientals whom he took to be Chinese and who greeted him affably. 
The strangers then suggested he and his party lay aside their guns. At this 
point the commander realized that he had been ambushed and gave the alarm. 
The Japanese machine guns opened on his trapped coiumn, inflicting heavy 
casualties. Some of his men made their way back to the Allied lines, but the 
company was never reconstituted and was broken up and distributed among 
the rest of Galahad. For his constant gallantry during a stubborn eight-hour 
rear-guard action, which permitted the survivors to extricate themselves from 
ambush, Pfc. Anthony Firenze of New Galahad received the Distinguished 
Service Cross. On 2 July reports came of the approach of a strong Japanese 
force from the north. Despite Stilwell's orders concerning the l/42d, Wessels 
thought his lines too thin and pulled back the l/42d, to strengthen himself 
toward the north. 113 

Hacking Out Small Gains 

The only gains in the week of 25 June-2 July were a few hundred yards 
taken by the 150th Regiment and the 236th Engineers. Monsoon rains, low 
visibility, and high water turned much of the terrain into a swamp in which 
men crawled, stumbled, waded, slipped, fell, and sometimes died. The few 
hundred yards that looked so small on the map were an immense distance to 
the men who had to crawl them under Japanese fire and keep them by beating 
off the inevitable Japanese counterattack. 114 Seen in retrospect, the prospect 
appears to have brightened slowly in the next few weeks. With every passing 
day experience improved the quality of the troops after the first shocks wore 
off. A training program was instituted on 7 July. Units in reserve were taught 
for eight hours daily; those in contact with the enemy trained four. The 3/88th 
and l/89th were put in reserve to prepare for a set- piece attack, to capture a 
stretch of the Sumprabum Road. 115 

Wessels' first attempt at a major attack was made on 12 July. It was a 
co-ordinated attack with the small amount of air support obtainable. Arranging 

113 (1) Siege of Myitkyina, pages 67-71, treats the episode very cautiously. (2) Wessels Interv 
cited n. 112(1). (3) Stilwell Diary, 29 Jun, 1, 2 Jul 44. (4) GO 13 1, Hq USAF CBI, 12 Oct 44. 
(5) A letter, Wessels to Ward, 3 October 1951, has as an inclosure a most interesting letter, 10 
July 1944, by the liaison officer with the battalion, Capt. Paul L. Tobey. OCMH. 

114 G-3 Wkly Per Rpt, Hq MTF, 25 Jun-2 Jul 44. NCAC Files, KCRC. 

115 G-3 Per Rpts, Hq MTF, 2-9 Jul 44; FO 17, Hq MTF, 1 1 Jul 44. NCAC Files, KCRC. 



this latter took time, and while the process was under way, a radio came from 
Stilwell: when is the attack of the 12th going to be made? A staff officer drew 
Wessels' attention to the hint in the radio, which he had missed, and the attack 
plan was hastily completed. Maj. Gen. Howard C. Davidson of the Tenth Air 
Force, who was consistently helpful and co-operative, on his own initiative 
arranged a bombing by thirty-nine B-25's. 

There was further support from the 88th Fighter Squadron. It gave 80 
percent of the fighter-bomber support at Myitkyina and was commanded by an 
officer who made a point of examining prospective targets from the front lines 
and consistently went beyond the call of duty in his efforts to give the infantry 
effective close-in support. From the beginning to the end of the siege, the Air 
Forces dropped 754 tons of bombs, or 20 percent more than the artillery 
effort. 116 

The attack on 12 July did not succeed. The intent was to have units in 
forward positions fall back while the bombing was done to avoid casualties 
from shorts, then advance immediately after. Air-ground liaison was ineffective; 
the B-25's underestimated the bomb safety line, and 40 percent of their bomb 
loads fell among the American troops north of Sitapur, causing some casualties 
and quite a bit of confusion. The gains by the 88th and 89th Regiments were 
again measured in hundreds of yards, and the attack simply petered out. How- 
ever, the series of attacks made since 17 May had its cumulative effect. The 
Allied lines were steadily constricting around Myitkyina. Units once separated 
by intervals through which Japanese seemed to slip at will were now in close 
contact. Of the Japanese garrison, 790 were dead and 1,180 wounded by 
mid-July. 117 

When the attempt at a co-ordinated attack bogged down, Wessels' men 
went back to their patient day-by-day advances, driving back the Japanese to 
their immediate front. On 21 July, in one of the patrol clashes that erupted now 
and again around the Allied strongpoints, Pfc. Marvin H. Dean, a Galahad 
veteran, won the Distinguished Service Cross for taking out a Japanese machine 
gun position that had stopped the patrol for which he was the lead scout. 118 

Stilwell came back on the 23d to check progress. The 149th Regiment joined 
the 50th Division on 24 July and took its place in line relieving portions of the 
42d and 150th Regiments. Next day the l/90th took over a quiet sector. These 
were substantial reinforcements. A day before they arrived, the first clear indi- 
cations came that the Japanese were relaxing their grip on Myitkyina. Eight 
rafts and a boat laden with Japanese were attacked on the Irrawaddy by Kachins 
of the OSS Detachment 101. Twenty-four Japanese were killed, two captured, 

116 (1) Wessels Interv cited n. 112(1). (2) JICA CBI Rpts 470, 477, 484, 6 Sep 44, sub: Air 
Support on Capture of Myitkyina. OCMH. 

117 (1) Maruyama Interrog; Interrog, Gen Honda, CG, 33d Army, 9 Jan 48. OCMH. (2) Siege 
of Myitkyina, pp. 84-85. (3) G-3 Per Rpt, Hq MTF, 9-16 Jul 44. NCAC Files, KCRC. (4) PW 

118 GO 131, Hq USAF CBI, 12 Oct 44. 



and then it was learned these were hospital patients fleeing Myitkyina. Three 
more Japanese seized by friendly Burmans revealed that hospital patients were 
being evacuated by the simple expedient of letting them drift down the river 
on rafts. 119 

On 26 and 27 July the 3d Battalion, New Galahad, waded across the 
crescent-shaped swale which had effectively protected the Japanese positions to 
the north and was finally on the northern airstrip which Mizukami's men had 
held so long. The 209th and 236th Engineer Battalions, which had now taken 
as heavy casualties as any American units in any theater (the former, 41 per- 
cent), were pulled out of line and put to defending the airstrip. In the last 
week of July the daily gains began to stretch out, and since the area held by the 
Japanese was steadily shrinking, began to reach deeper into their vitals. Japa- 
nese counterattacks were no longer so dynamic, the positions captured from 
the enemy were no longer so well made, and many Japanese dead were found 
to be badly wounded men returned to the line. As July ended, gains of several 
hundred yards a day were frequent, though still costly. On the 28th, Galahad 
veterans T/5 Russell G. Wellman and Pfc. Herman Manuel teamed to rescue a 
wounded comrade from under "intense enemy machine gun and rifle fire." 
Though they were wounded, Manuel and Wellman succeeded in their gallant 
attempt, and won the Distinguished Service Cross. 120 

One of the siege's command problems was uncontrolled and seemingly 
uncontrollable fire by the Chinese. With all ammunition coming in by air, 
every wasted round was to be deplored, yet time and again Chinese units let fly 
with all they had, and in every direction. The night of 30 July apparently was 
especially memorable in this regard, and General Wessels ordered an investiga- 
tion. In one Chinese battalion alone, ten men were wounded by this wild firing. 
And one Chinese interpreter, though not classed as among the most reliable, 
stated that the 89th Regiment was so angered by the firing that "it was ready 
at any time to take on the 4 1st!" The report concluded: 

4. All of the above-listed Liaison Officers are willing to testify, under oath, that their 
lives were endangered by the throwing of hand grenades and by firing of automatic weapons 
going in 3 directions, over and near their positions during the night; that the throwing of 
hand grenades and firing was done by the Chinese. 121 

The Last Days 

Within Myitkyina, General Mizukami made his last decisions. At the end 
of July, Colonel Maruyama requested that his regiment be withdrawn. Mizu- 

119 Siege of Myitkyina, pp. 105-09. 

120 GO 131, Hq USAF CBI, 12 Oct 44. 

,21 Ltr, Col Laughlin, ACofS, G-3, MTF, to Wessels, 31 Jul 44, sub: Rpt of Firing During 
Night 30-31 July. Wessels File, OCMH. 



kami's intention had been to fight to the last, but he agreed. Instructions were 
issued accordingly, and Mizukami, having done his duty as a soldier, made his 
apology to his Emperor and committed suicide on 1 August. 122 This was, of 
course, unknown to Wessels' headquarters, but the weakening of Japanese 
resistance had become ever more obvious. Attacks on the 29th, 30th, and 31st 
made ever deeper advances into the Japanese lines. Capts. Shields A. Brubeck 
and John J. Dunn and 1st Lt. Donald W. Delorey, all of New Galahad, led 
their units during these days with valor that brought them the Distinguished 
Service Cross. 123 

Weighing the many evidences that control of the situation was rapidly 
passing into their hands, Wessels and his colleagues drafted a new plan of 
attack. It included an ingenious device, credited to General Pan Yu-kun of the 
50th Division. A raiding party, formed into fifteen heavily armed sections, was 
organized and briefed on infiltrating Japanese lines facing the 50th Division. 
Having made its way through the Japanese lines in darkness, it was to remain 
hidden until 0430 when the 50th would assault while the raiders spread confu- 
sion behind the Japanese lines. Meanwhile, air reconnaissance revealed many 
rafts moored against the Irrawaddy within the Japanese lines. 124 

"At 0300 hours (3 August), when the moon went down and rain and 
thunder set in, the raiding party of the 50th Division moved out. The approach 
was detected only once and some shots were fired at the raiders, who hit the 
ground. The Chinese did not return fire however, and after laying low for a 
while, moved out again quietly and cautiously." Later the 50th Division 
attacked and, thanks to the efforts of the raiders and the preliminary stages of 
the Japanese evacuation, made rapid progress. The attack spread rapidly up the 
Allied line as regiment after regiment moved forward with comparative speed. 
Resistance by the Japanese rear guards was speedily overcome and at 1545, 
3 August, Myitkyina was officially called secure. One hundred eighty-seven 
Japanese prisoners were taken, most of them patients. Colonel Maruyama made 
good his withdrawal with about 600 men. 125 

The direct cost of Myitkyina to the Allies was: 

Killed Wounded Sick 

Chinese 972 3,184 188 

American 272 955 980 

Some of the Chinese shown as killed may well have been among the deserters 
who after the war created something of a problem in the Myitkyina area; others 
probably perished in the fighting among Chinese units during the first confused 

122 Japanese Officers' Comments, Incl 6, Col Masanobu Tsuji. Mizukami's adjutant was among 
those evacuated, and reported later to Colonel Tsuji. 

I2 ' (1) GO 131, Hq USAF CBI, 12 Oct 44. (2) GO 141, Hq USAF CBI, 24 Oct 44. 

124 Siege of Myitkyina, pp. 128-29. 

125 (1) PW History. (2) Tanaka Interrog. OCMH. (3) Siege of Myitkyina, pp. 130-32. 



attempts to take Myitkyina. Of the U.S. sick, 570 were from Galahad; 126 even 
so, the discrepancy between the Chinese and American figures shows the virtues 
of the Chinese insistence on drinking only boiled water and eating only cooked 
food, as well as the very considerable resistance of the Chinese to a 
contaminated environment. 

The attack had been costly in suffering and losses, but the Mogaung- 
Myitkyina area in Allied hands was a great prize and marked an achievement 
of which the end of the fighting at Myitkyina was the outward and visible 
symbol. In the first instance, it helped increase Hump deliveries to China. As 
the fighting moved ever southward down the Hukawng and Mogaung valleys, 
it became safer for ATC transports to use the lower and more southerly routes 
to China. When Stilwell's men stood on the main airstrip at Myitkyina on 
17 May 1944, the wider, southern route over lower terrain was a reality. Almost 
simultaneously with the taking of the airstrip the India-China Wing of ATC 
began to reap the benefits of better maintenance and a steadily increasing allot- 
ment of aircraft. The combination of more and better transports, better main- 
tained, flying a shorter, lower, safer route was a potent one, and was speedily 
reflected in the Hump tonnage deliveries, 127 whic h rose from 13, 686 tons in 
May to 18,235 tons in June, and 25,454 tons in July fffig Chart ^.j| That in the 
months to come Hump tonnage would be of such an order of magnitude meant 
a great change for the better in the American position in China. American 
officers there would in the future have resources that Stilwell and Chennault in 
1942 and 1943, worrying over the distribution of a few thousand tons, would 
have regarded as sheer opulence. 

In the second place, occupation of the Mogaung-Myitkyina area meant that 
as soon as the pipeline and Ledo Road reached it, the Allies would have a great 
supply base squarely on the road and rail net of Burma, and within easy dis- 
tance of China itself. About 1 May 1944, as noted above, Stilwell had concluded 
that opening the line of communications to China was not within his capabili- 
ties and had so informed his superiors. Then the results of Slim's and GifFard's 
decision to meet and break the Japanese 13th Army at Imphal began to be 
apparent, together with the heavy casualties the Japanese were suffering around 
Kamaing, Mogaung, and Myitkyina. Observing this, Stilwell changed his mind 
about the possibility of taking the key points which controlled the line of 
communications. He told Mountbatten in mid-July that the Japanese had 
suffered very heavy losses, that it was doubtful if by November they could 
recover two thirds of the effectiveness they had possessed in early 1944. He was 
convinced that the Allies could now do better than just hold Myitkyina, and so 
he assured Mountbatten that "with reasonable help from the Y-Force, the CAI 
can get to the Bhamo-Shwegu area." The next step would be to take Lashio, 

126 Siege of Myitkyina, pp. 130-32. 

127 History of the India-China Division, Air Transport Command, I (1944), 39. Hist Div, 
MATS, Andrews Field, Md. 



terminal of the Burma Road and the prewar gateway to China. Plainly, to 
Stilwell the end of China's isolation was near. 128 

Stilwell's renewed optimism was in complete accord with conclusions of 
Slim and the latter's corps commanders. After visiting them at the front at this 
same period of mid-July, Mountbatten recorded in his diary that "the thing 
that struck me most was the absolute certainty that whatever else happened 
we must start our offensive after this monsoon before the Japanese can start 
theirs. . . ." The question that remained was to plan the offensive whose 
execution would be begun in the fall of 1944. 129 

When it became apparent that Myitkyina was about to fall, Stilwell left 
north Burma for SEAC's headquarters at Kandy, Ceylon. Mountbatten was 
going to London to rearrange SEAC's command structure, and Stilwell was 
happy that his position as acting Deputy Supreme Allied Commander made 
it obligatory for him to visit the beautiful island and enjoy a brief respite from 
field command. 130 Since there were an army group and an army headquarters 
to handle the land fighting, and since no major naval operations were con- 
templated, Stilwell was glad to let SEAC headquarters run itself while he 
toured the island and rested. 

The press took the opportunity of examining him on the North Burma 
Campaign, and Stilwell on 5 August discussed it with candor: 

We think there were approximately 1,000 [Japanese] in Myitkyina, when we struck it. 
Various units got into Myitkyina because our cordon around the town had holes in it. Cer- 
tainly more got in there than I calculated, and I admit underestimating the strength of the 
Jap garrison. The situation was very confused. 131 

In Stilwell's opinion, the Chinese soldier best withstood the hardships of 
the campaign. He saw no change in the high quality of the Japanese soldier. 
Take a man from any Japanese service unit, said Stilwell, and he will get into a 
hole with a light machine gun and stay there. Stilwell believed that the 18th 
Division was destroyed. He did not know of any units that might have escaped 

128 (1) History of India-Burma Theater, 1944-1945, I, 81, 173. OCMH. (2) Memo, Stilwell 
for Mountbatten, 18 Jul 44. SAC (44) 288. SEAC War Diary. (3) The pipeline reached Myitkyina 
on 2 October 1944, and a temporary combat road was opened to Myitkyina in early November. 
The Ledo Road itself bypassed Myitkyina. 

129 (1) Personal Memo for C-in-C's on SAC's Visit to Burma Front, SCM/44, 6 Jul 44. SEAC 
War Diary. (2) Allied Burma operations of 1944-45 are described in Volume III of this subseries. 

1,0 (1) The Stilwell Papers, p. 310. (2) When, if ever, Stilwell was formally appointed Deputy 
Supreme Allied Commander is not fully clarified by the sources the authors have been able to con- 
sult. In March 1944, drafting notes for a conference with Mountbatten, Stilwell wrote that he had 
never been appointed Deputy. See page 169, above. A lengthy (seventeen-page) handwritten list 
of differences and difficulties between Stilwell and SEAC, which internal evidence indicates was 
written very near the close of Stilwell's stay in CBI, was found by Sunderland at Carmel, Calif., in 
May 1950, and filed with the JWS Miscellaneous Papers, 1944. It contains the statement that SEAC 
never issued orders appointing Stilwell as Deputy. On the other hand, Mountbatten' s report always 
refers to Stilwell as Deputy. When Stilwell assumed command at Kandy in July 1944 it was his 
first exercise of authority as a Deputy, and he seems to have confined himself to cutting down the 
flow of minutes and staff studies (paper work of all kinds being one of Stilwell's phobias) and to 
presiding over meetings. 

131 Interv of Stilwell by press, Kandy, Ceylon, 5 Aug 44. SNF 172. 



intact. A few men from the 55th and 56th Regiments might have escaped, but 
he thought the 114th Regiment was gone, together with the division artillery 
and vehicles. Stilwell did not know how many aircraft the crisis at Imphal had 
diverted from the Hump but agreed that of course it had cut Hump tonnage. 

That turned the conversation to China and to the situation there. It was 
grave, Stilwell conceded. Because of his official position he could not comment 
bluntly and said as much. What he did say was tactful and sympathetic toward 
the Chinese in their time of sorrow. China was his next problem, and he was 
soon to go there. 


The taking of Myitkyina on 3 August, which successfully accomplished a 
task many informed observers had termed impossible, marked a milestone in 
the history of CBI. It could not be termed a successful carrying out of Stil- 
well's plans, for Stilwell's own plan, which he presented in the summer of 
1942, called for a drive from Imphal with the Chinese Army in India, together 
with a Chinese drive from Yunnan, aimed at taking Rangoon and reopening 
the prewar line of communications from Rangoon to Kunming. These oper- 
ations, Stilwell thought, should be launched during the dry season that began 
in November 1942 and be completed sometime in 1943. 

Stilwell's superiors, in successive conferences, whittled down some of these 
conceptions and substituted others of their own, so that all the final plans had 
in common with Stilwell's own proposals was that both envisaged a land 
campaign in Burma. The manifold obstacles, political and military, that had 
to be surmounted before the campaign opened postponed it until fall 1943, or 
one year after the time Stilwell had sought to begin it. 

But though Myitkyina was taken very late in the day, taken it was, and the 
feat was a triumph for the man who had maintained it could be done. The 
town was taken to make possible an intensified air effort from bases in China 
in support of U.S. operations in the Pacific and it is against subsequent events 
in China that the Burma campaign, and the decisions that resulted in post- 
poning it from 1942 to 1943, must be weighed. 


Logistics and Administration 

Behind the successes in north Burma and Manipur State lay successes in 
the field of logistics. Probably the most important of these was the improve- 
ment in the Assam line of communications that followed its passing under 
Anglo-American military control. There were also major administrative 
changes in both SOS and theater organization. These may have contributed 
to victory in the field, though with less obvious directness than did the in- 
creased flow of supplies to the front. 

The Headquarters Reorganized 

In India, responsibility for American supply, maintenance, and construc- 
tion operations lay with the Commanding General, Services of Supply, save for 
a few exceptions. In China, the situation was extremely complex and is treated 
separately below. In CBI after 15 November 1943, General Covell com- 
manded the SOS, while his predecessor, General Wheeler, became Principal 
Administrative Officer, SEAC. 1 General Covell, after surveying his new com- 
mand in the light of the added resources that would come to CBI as a result 
of the Quadrant decisions and give him a freedom of action Wheeler had 
never enjoyed, suggested certain major changes. 

He proposed to reorganize the SOS along lines suggested earlier by the 
War Department to meet the world-wide problem of an efficient division of 
responsibilities between theater headquarters and SOS. Theater headquarters 
approved generally, but preferred to keep Signals, Postal Service, Censorship, 
Special Services, Military Police, and Malaria Control under its own jurisdic- 
tion. \£TaH£4)_ Covell at once moved to set up general depots and Engineer 

and Transportation Services under SOS control. The American port activities 
at Bombay became an exempted station under the Chief of Transportation, 

General Covell wanted to simplify the geographical setup by merging 
Advance Section No. 1 (Gaya) with Base Section No. 2 (Calcutta,) but hesi- 
tated for a while because Calcutta was not working to his satisfaction. Covell 
also wanted to merge Advance Section No. 2 and Base Section No. 3 into an 

1 Rpt, Covell to CG, USF IBT, 20 May 45, sub: Final Rpt, 15 Nov 43-20 May 45, pp. 1-2. 
(Hereafter, Covell Report.) OCMH. 



Table 4 — Strength of U. S. Army Forces in the China-Burma-India 
Theater: January-September 1944 

End of month 



Services of 

Army A 

Air Transport 

r Forces 



lUi, U/J 

21, 064 

"27 1 CI 

11, 5oi 




19, 835 

42, 574 


42, 926 





47, 239 

12, 202 

52, 054 

2, 546 


154, 521 


48, 005 


72, 997 



162, 506 


50, 355 

12, 974 

77, 214 


169, 111 

22, 206 

51, 707 


80, 787 



175, 546 

22, 436 


14, 102 


2, 042 


186, 364 

24, 245 

53, 186 

16, 160 

91, 586 

1, 187 


188, 565 



19, 133 

90, 825 


• Data not reported. 

Source: G-4 Personnel Rpt, quarter ending 30 Sep 44, Hq USAF CBI. OCMH. 

Intermediate Section, but being anxious not to interfere with Colonel Pick, 
who commanded Base Section No. 3, and the road-building effort, he con- 
templated waiting until Pick was ready to displace forward as the Ledo Road 
progressed. 2 

In his own headquarters, Covell set up a chief of administration and a chief 
of operations, dividing the general and special staff functions between them. 
His aim was to reduce the number of officers reporting to him. Ultimately, 
SOS reverted from this organization to the customary general and special staff 
organization. 3 

Once the Engineer Construction Service had been set up under Col. 
Thomas F. Farrell, SOS chief engineer, Engineer Division No. 1 (Base Section 
No. 1) and Engineer District No. 2 (Advance Section No. 2, Base Section No. 
3, exclusive of the Ledo Road), plus Engineer Division No. 3 for the Delhi- 
Agra area, were organized to take care of all the SOS major engineering 
projects. 4 

Possibly feeling that with Covell changing the SOS the time had come for 
a renovation of the theater's organization, General Hearn on 17 January sug- 
gested to General Sultan, Deputy Commander, CBI Theater, that theater head- 
quarters itself should be established in New Delhi, and that Ch ungking head- 
quarters should be redesignated Headquarters, Forward Echelon. (See Chart 2.) 

2 (1) Ltr, Covell to Somervell, 16 Dec 43. Somervell File, Vol III, Hq ASF, Ts of Opns, CBI 
1942-43, A46-257. (2) Roland G. Ruppenthal, Logistical Support of the Armies, Vol. I, Chs. I, III, 

3 (1) SOS in CBI, p. 24. (2) GO 16, Hq SOS USAF CBI, 1 Feb 44. (3) GO 114, Hq SOS 
USAF CBI, 11 Aug 44. 

4 Ltr, Covell to Somervell, 27 Jan 44. Somervell File, Vol IV, CBI 1944. 



This would result in an administrative structure more closely related to exist- 
ing circumstances. One circumstance was that since Stilwell had spent most 
of his time in India and Burma, it would be appropriate for his principal head- 
quarters to be on that side of the mountains. Another was that because of 
geographic barriers, most of the American men and material in CBI were in 
India and Burma, and there of necessity were the bulk, even if not the most 
complex, of the theater's administrative problems. For Stilwell's brief visits to 
Chungking, a small headquarters would serve very well. 5 

For whatever reason, Sultan did not take up Hearn's suggestions with Stil- 
well until March 1944. Stilwell promptly approved them. The general order 
which announced the change on 31 March 1944 made plain that there was a 
certain shift in the center of the theater commander's interest away from China. 
Theater headquarters at New Delhi was made responsible for "formulating 
general overall policies and announcing them to the Command, for planning 
and for the administration of the Theater. ..." Forward Echelon headquarters 
was given strictly limited responsibilities: 

a. Liaison and coordination with the Chinese Government on all matters of common in- 
terest to the United States Army Forces and the Chinese Government. 

b. Securing and transmitting information from China as required by instructions from 
Theater Headquarters. 

c. Coordinating and supervising the execution within China Theater of orders, directives, 
et cetera. 

d. Assisting Theater Headquarters in the preparation of policies, conduct of planning, 
and establishment of procedures concerning the operation and the administration of United 
States Army Forces and supporting installations in China. 6 

When the order was put into effect, word of the change quickly spread 
through higher Chinese social and governmental circles in Chungking, in the 
distorted form that the United States Army was withdrawing from China. U.S. 
Ambassador Clarence E. Gauss was disturbed by the rumor and took the mat- 
ter up with Hearn. Hearn quickly reassured him, but the fact that reassurance 
was necessary suggested that liaison between the Embassy and theater head- 
quarters was not what it might have been. 7 

SOS Problems 

On surveying his far-flung command, Covell noted that most of his prob- 
lems came from two factors not under SOS control, the port of Calcutta and 
the Assam line of communications. Despite the firm conviction of local author- 

5 (1) Stilwell's Mission to China J"Ch*"lH (2) Rad CFB 12494, Hearn to Sultan, 17 Jan 44. 
Item 1635, Bk 5, JWS Personal File. 

6 GO 5, Fwd Ech, Hq USAF CBI, 3 1 Mar 44. 

7 (1) Rad CFB 15614, Hearn to Marshall, 2 Apr 44. Item 2160, Bk 6, JWS Personal File. (2) 
CM-OUT 18999, Marshall to Hearn, 5 Apr 44. (3) Ltr, Gauss to Hearn, 2 Apr 44; Memo, Hearn 
for Gauss, 3 Apr 44; Ltr, Gauss to Hearn, 12 Apr 44. Items 292, 294, Bk 3; JWS Personal File. 



ity that the Assam line of communications could not be improved in time to 
support the projected operations in Burma, Covell had been told by Somervell 
that the line could and must be substantially improved in a matter of weeks. 8 
The first problem Covell proposed to attack was that of the line of communica- 
tions. Solution of this would permit the more rapid flow of supplies to SOS's 
great engineering projects, and other major projects— the Ledo Road, the pipe- 
lines, and the airfields. It would also permit supplies in quantity to reach the 
British forces on the Manipur and Arakan fronts, whose supply problems had 
been so severe that in some areas troops had suffered from malnutrition. 9 

The most immediate problem, and the one with aspects of emergency, was 
the supply situation in Assam. Because low water on the Brahmaputra River 
impeded barge operations, and because the meter-gauge lines of the Bengal and 
Assam Railway were clogged with traffic, stocks in Assam dwindled to a seven- 
day level. Since it then required sixty-seven days for supplies to go by rail from 
Calcutta to Assam, it was possible to foresee a situation in which supplies 
would be below the subsistence level, and in which timely replenishment 
would be most difficult. 10 

To forestall this emergency, and to impress upon local authority the grave 
situation resulting from the inefficient and lethargic operation of the line of 
communications, requests were made for several complete trainloads of speci- 
fied items to be made up and shipped to Assam on passenger-train schedules. 
The Indian rail authorities and the Bengal and Assam Railway agreed, and the 
emergency passed. 11 


Negotiations with the British to arrange military control of the port of 
Calcutta and the line of communications began in a favorable atmosphere, 
with the British and American military authorities soon finding themselves in 
accord as to what had to be done to improve matters. Both sides wanted a 
British port controller with a U.S. assistant to regulate and control port facili- 
ties and personnel; definitely assigned dock areas to permit each party to 
concentrate its resources in its area; a central pool of dockside labor; pooling of 
lighters. 12 Putting a British officer with an American assistant in charge of the 
entire line of communications seemed desirable to both parties. 

The British and American supply authorities who engaged in the pre- 
liminary talks secured the assent of CBI and SEAC to their proposal for a port 
controller and military control of the entire line of communications from 

8 Covell Report, pp. 1-2. 

9 Ltr 3872/17/Q, Slim to GHQ (India), 30 Oct 43. SEAC War Diary. 

10 Covell Report, pp. 27-28. 

11 (1) Ibid. (2) SOS in CBI, pages 425-26, has an excerpt from the Covell Report in a dis- 
cussion of the general difficulties of supplying U.S. installations in Assam. 

12 (1) Ltr cited n. 2(1). (2) Ltr, Covell to Wheeler, 16 Jan 44; Ltr, Covell to Somervell, 1 Feb 
44. Somervell File, Vol IV, CBI 1944. 



Calcutta north. The lower echelons of the Government of India approved in 
principle. Then the proposal seemed to encounter heavy weather. While SOS 
waited for clearing skies, it urged Army Service Forces to have a supremely 
well qualified man ready to represent the United States as assistant port 

By 21 January, Covell was ready to write: 

Probably our biggest headache has been with the L of C from Calcutta to Assam which 
is in British civilian hands. We get all kinds of promises but nothing happens. As a result, 
for the last two weeks I have been raising hell, but still feel that more drastic action may be 

In December with an approximate equal amount of tonnage allocated to the U.S. and 
British military, the U.S. Army lost 14,981 tons while the British lost only 723 tons. As a 
result of our vigorous protests we received the promise that this shortage would be made up 
in January 1944 and the deficiencies divided equally thereafter. Now we have just learned 
that the line is congested and we still will not get what was promised to us. The whole 
situation looks bad to me, and I am having great difficulty in digging out necessary informa- 
tion. I get the impression that the Indian Civil Railway Administration is not much con- 
cerned about military traffic. I have been trying for two months to find out reliable total 
figures over the broad gauge road north of Calcutta, but have been unable to get any true 
figures on civilian traffic, but I am still trying. 13 

The drastic action Covell wanted was immediately forthcoming. General 
Marshall put the matter before the President. Describing the situation on the 
line of communications as "precarious," and pointing out that three of the ATC 
airfields in which the President took such close interest were out of gas, 
Marshall said that Anglo-American military control of the line of communica- 
tions was the only solution, and asked that Roosevelt write to the Prime 
Minister. 14 The President agreed and, pointing out that the operations of the 
ATC had been embarrassed by lack of vigorous management on the line of 
communications and that the results of civilian administration had been dis- 
appointing, he urged that Churchill intervene in person to obtain full military 
control of the line of communications from Calcutta to Assam. In reply, the 
Prime Minister promised a personal investigation. 15 

A few weeks later, SOS learned that the War Transport Department of the 
Government of India had been strongly opposing any move to place the port 
of Calcutta under military control. The civilian authorities wanted a com- 
missioner who had been prominently identified with the existing regime to 
have charge of the port. The military, British and American, vigorously 
opposed the suggestion. They swept the field when the Viceroy, that distin- 
guished soldier, Field Marshal Lord Archibald P. Wavell, decided personally 

13 Ltr, Covell to Somervell, 21 Jan 44. Somervell File, Vol IV, CBI 1944. 

14 Memo, Marshall for President, 29 Jan 44, sub: Failure of Calcutta-Assam Line of Commu- 
nications. Somervell File, Vol IV, CBI 1944. 

" Col. Charles K. Gailey, Exec Off, OPD, for Handy, 31 Jan 44. Item 55, OPD Exec 10. At- 
tached to the memorandum is a draft radio, Roosevelt to Churchill, which Colonel Gailey says 
was dispatched on 29 January. Colonel Gailey also reports Churchill's 30 January reply to the 



that there would be a complete change in Calcutta. Perhaps remembering how 
his own operations in Burma had been hobbled by logistical problems that 
were called insoluble, the Viceroy, "with the accompaniment of a little desk 
pounding," made his wishes known. The civilian departments heard and 
obeyed. 16 

The agreement on the line of communications that resulted on 6 February 
provided that an outstanding man should control the entire working of the 
port, and all agencies concerned with it. Under him would be two deputies, one 
of them an American. To meet civilian objections, it was agreed the port 
director would be a civilian, and responsible to the War Transport Department 
of the Government of India. 17 

Another agreement, completed the same day, provided a new set of controls 
for the line of communications. Responsibility was placed on the Calcutta 
representative of General Headquarters (India), advised by the regional con- 
troller of traffic priorities for the Government of India, a representative of the 
U.S. Army, and representatives of the railway and barge lines. 

3. The mechanism for implementing the above decisions will be as follows: 

(a) The total capacity of the L of C will be estimated periodically by D. D. Mov., 
Calcutta [Deputy Director, Movements], in conjunction with the Railway and Steamship 
Companies concerned, and in consultation with the Officer Commanding U.S. Railway 
Troops. This estimate of capacity will be telegraphed immediately to PAO (I) [Principal 
Administrative Officer, India], and PAO (SEAC). 

(b) Proposals as to the allotment on each sector and in both directions of this capacity 
as between military, essential civil, and railway construction and maintenance requirements, 
and the routing of three tonnages, will be prepared in Calcutta . . . by . . . 


Representative of Commanding General, SOS 
Representative of Army Group, S.E.A. 
Regional Controller of Priorities, Calcutta north 
Representative of B & A Railway, 

and submitted to Delhi where they will be agreed or amended by representatives of GHQ 
(I), War Transport Department, H.Q., S.A.C., S.E.A.C. and the Railway Board, the 
approved constitutional procedure being followed in the event of failure to agree. 

(c) Responsibility for implementing the agreed allotments of (b) will be placed on a 
panel constituted as in (d) below. . . . 

(d) The panel . . . will consist of D. D. Mov., Calcutta, who will be chairman . . . 

Regional Controller of Priorities, Calcutta north 

Commanding General, S.O.S. 

B & A Railway 

Rivers Steam Navigation Co. 

India General Navigation Co. 

Movement Control (Railway) 

Movement Control (IWT [Inland Water Transport]) 18 

l6 Ltr, Covell to Somervell, 11 Feb 44. Somervell File, Vol IV, CBI 1944. 

17 Incl to Ltr cited n. 16. 

18 Ltr, Ferris to Marshall for OPD, 11 Feb 44, Incl, sub: Calcutta Port and Assam LOC. Somer- 
vell File, Vol IV, CBI 1944. 



THE KING GEORGE DOCK AREA in Calcutta, India. 

Clearing the Port of Calcutta 

Clearing the congested port of Calcutta was unwittingly facilitated by the 
Japanese. Submarine activity in the Bay of Bengal during December 1943 
forced the convoying of ships from Ceylon to the mouth of the Hooghly River, 
upon whose banks Calcutta is built. This threat slowed the arrival of cargo and 
permitted the dispatch of much that had accumulated on the docks and in the 
warehouses. For the Americans, the burden fell on two port companies of 
Negro soldiers, the 540th and the 54 1st (Transportation Corps). For months 
these men had worked like Trojans, sometimes right around the clock. 19 

During the winter months, British consent to SOS control of part of the 
great King George Docks was obtained. In the week of 29 December 1943- 
5 January 1944, seven companies (the 497th and 508th Port Battalions, TC) of 
veteran longshoremen with complete equipment for unloading ships arrived. 
In the month of January, the ten U.S. port companies then on duty unloaded 
98,859 long tons, as against 42,325 long tons the month before. 20 

19 (1) SOS in CBI, App. 3, History of Base Section No. 2. (2) Ltr, Covell to Somervell, 4 Jan 
44. Somervell File, Vol IV, CBI 1944. 

20 App. cited n. 19(1). 



In addition to the tripling of the U.S. force at hand and the leasing of the 
King George Docks, other steps contributed greatly to faster movement of 
cargo through the port. 

1. The procurement of large floating cranes and other modern dock equipment. 

2. The diversion of barge line equipment to augment port facilities. 

3. The partial replacement of native labor with troops having stevedoring experience, the 
training of other units along these lines and the pooling of native labor. 

4. The establishment of a 24-hour work day with shifts to maintain it. 

5. The more efficient use of transportation facilities in removing cargo from the docks. 

6. The improvement in the Assam LOC. 21 

On 15 March 1944 the King George Docks were completely cleared for the 
first time. During the next thirty days the dock workers handled 97,000 
measurement tons of freight. By 20 April, General Co veil could report: . . no 
special concern need be given to the capacity of the port of Calcutta, nor has 
any indication been given at this time that it has reached its saturation point." 22 
The port director, Mr. F. A. Pope, arrived 22 May 1944. 

On 23 January Brig. Gen. Gilbert X. Cheves was assigned to the command 
of Base Section No. 2, succeeding Brig. Gen. John A. Warden, who went to the 
west coast of India to take command of Base Section No. 1 (Karachi). General 
Cheves found a problem in relations between U.S. troops and the people of 
Bengal. SOS personnel had not been well disciplined. The port of Karachi 
under General Warden continued to work with smooth efficiency. Its rate of 
discharging cargo never ranked below fifth among 110 ports in which the U.S. 
Army operated. 23 

The steady flow of reinforcements to CBI after the Quadrant Conference, 
August 1943, made it desirable to bring troops in through Bombay. For one 
reason, some of the transports were too large to use the Karachi facilities, and 
for another, the rail connections out of Bombay were better. In 1943, 118,983 
U.S. troops passed through Bombay, some of them intended for the Persian 
Gulf Command. 

Local authorities at Bombay were most co-operative in finding office space, 
hospital accommodations, and staging area quarters for American use. As a rule, 
troops remained on board their transports until the troop trains appeared. If it 
was necessary for the ship to leave before the trains came, then the staging 
areas were used. The principal drawback to this arrangement lay in the fact that 
British rations as prepared by Indian kitchen personnel were not satisfactory to 
American troops. Some relief was found in supplying the difference between 
the two rations, close supervision of the kitchens, and the partial use of 

21 Covell Report, p. 8. 

22 Ltr, Covell to Somervell, 20 Apr 44. Somervell File, Vol IV, CBI 1944. 

23 (1) GO 10, Hq SOS USAF CBI, 23 Jan 44. (2) SOS in CBI, p. 35. (3) Rpt, Covell to 
Somervell, 26 May, 21 Jun 44. Somervell File, Vol IV, CBI 1944. 



American mess personnel. An American staging area, initiated in February 
1944, received its first troops on 24 July. 24 

Railway Problems 

Restricting factors that affected tonnage moving up over the railway lines 
of the Bengal and Assam Railway lay both in the physical structure of the line 
itself and in the methods of those who operated it. The railway was partly of 
broad gauge and partly of meter. The broad-gauge line seemed capable of haul- 
ing more traffic than the meter-gauge line could receive; bottlenecks lay in the 
transshipment points, where coolies lethargically carried stores from broad- to 
meter-gauge cars and vice versa. There was no bridge across the Brahmaputra 
River, and the ferry between Amingaon and Pandu was an obstruction. 
Together with the stretch of steep-gradient track between Lumding and 
Manipur Road (near Dimapur), the ferry was regarded as a major obstacle. 
j(See Map iTj 

Remedial construction was under way. The double-tracking of the Lumding- 
Manipur Road section had been begun by Indian authority but the pace of the 
work drew blistering comment from American railwaymen. Between July 1943 
and April 1944, 3.25 miles had been completed. 25 The Government of India had 
made plans for a bridge across the Brahmaputra River, for double-tracking the 
main line, and for new yard facilities. However, such projects would require 
years to complete and would themselves be a major burden on the line of 
communications. 26 

Signal communication along the Bengal and Assam Railway was lacking. 
Daily traffic figures were therefore not available, and trains would be "lost" 
four or five hours at a time. 27 

Operating methods seemed strange to American railwaymen. The railway 
authorities of the Government of India thought it possible to analyze the exist- 
ing facilities of a line of communications and from them determine its capacity. 
This estimate in turn would be set as a target. When the target had been 
reached, the mission would be regarded as accomplished. Moreover, the operat- 
ing echelons on occasion took the target figure as literally indicating the limit 
of the line and refused to go beyond it. 28 A major element in attempts to set 
target figures was the local belief that railway management had to keep the 
number of loaded cars moving up to the front equal to the number of empties 
returning from the front. The American approach, on the other hand, con- 

24 SOS in CBI, pp. 35-36. 

25 Ltr, Col Appleton, Director, MRS USAF CBI, to Brig Gen Thomas B. Wilson, CG, Trans- 
portation Service, 4 Apr 44. Somervell File, Vol IV, CBI 1944. Appleton thought there was a 
deliberate slowdown. 

26 Ltr, Appleton to Wilson, 25 Apr 44. Somervell File, Vol IV, CBI 1944. 
r Ltr cited n. 25. 

28 (1) Ltr, Covell to Somervell, 12 Apr 44. Somervell File, Vol IV, CBI 1944. (2) Ltr cited 
n. 25. 



centrated on moving loaded cars to the front and accepted the resulting lack of 
balance until the returning empties corrected the situation. To American rail- 
roadmen, the quantity movement of freight was most important; to Indian 
railroadmen, the careful balancing of loaded and empty freight cars to avoid 
even a temporary drain on the resources of other lines. 29 

Military Railway Service Begins Its Work 

Anticipating a share of responsibility in operating the Assam line of com- 
munications, SOS established the Military Railway Service on 25 December 
1943, under the Chief of Transportation, SOS, CBI. The main body of 
American railway troops whose assignment to India had been agreed to in the 
fall of 1943, arrived in India on 11 January 1944. 30 Advance elements of the 
railway troops had meanwhile engaged in reconnoitering the sections of line 
that the U.S. was to assist in operating. Officials and officers of the Government 
of India and of SOS framed a series of agreements under which the American 
railwaymen would function. 

The basic principle of the agreements was that American personnel to the 
number of about 4,600 would be superimposed on the existing railway staff. 
No employees of the Bengal and Assam Railway were to lose their jobs. The 
Military Railway Service took over operation of 804 miles of main and branch 
lines. At Parbatipur, the Military Railway Service transferred its traffic between 
the broad- and meter-gauge lines and operated the bottleneck ferry at Pandu. 
Military Railway Service did not affect commercial traffic which continued to 
be completely under the Bengal and Assam Railway. 

On the administrative side, the Military Railway Service and the Bengal and 
Assam Railway were each responsible for discipline of their own personnel. 
The railway management undertook to supply all expendable railway stores. 
The current construction program continued. The Military Railway Service 
agreed to ask for new construction through the existing railway channels. 31 

Headquarters of the Military Railway Service was set up at Gauhati, a 
central location and close to the two principal bottlenecks, the Pandu ferry and 
the Lumding-Manipur section of track. Headquarters personnel were members 
of the 705th Railway Grand Division. Colonel Appleton, formerly general 
manager of the New York Zone of the Pennsylvania Railroad, commanded. 
The operating battalions deployed, and the Americans took up their work at 
0001, 1 March 1944, on the eve of the Japanese invasion of India. 32 The relation 
between the two events was a coincidence, but it is interesting to speculate on 
what might have happened had the Assam line of communications been unable 
to meet the responsibilities soon thrust on it. 

29 Ltr cite d In 25l 

30 SOS in CBI, pp. 23, 63. 

31 SOS in CBI, pp. 90-93, Summary of Agreements for Opn of Portions of B&A Ry by USAF, 

32 SOS in CBI, pp. 63-66. 



Changes in Operating Procedure 

Colonel Appleton at once moved to break the bottleneck at the Pandu 
ferry. Work was begun immediately by using two locomotives simultaneously: 
one engine moved cars off the barge, and another, already coupled up, was 
ready to move cars on the barge. Prior to taking over, the Military Railway 
Service had agreed to a daily target of 305 cars eastward over the ferry; a daily 
average of 327 cars was achieved. A movement of 350 cars plus construction 
trains was projected for April. By 21 April the daily average was up to 413 cars 
eastward, and still rising, as shown by the 540 cars moved on that day. 33 

For Parbatipur, the transshipment point from broad to meter gauge, 
Appleton directed that the target figures be dispensed with, that cars be loaded 
as fast as they were available. Appleton's directive plus the improvement of the 
ferry operations meant that an increased number of loaded cars began moving 
east. An increase in the operating efficiency of the railroad was called for. 
Changes in operating procedure provided this. Train lengths were doubled, 
permitting more efficient use of motive power. Hundred-car trains were not 
uncommon. Train speeds were increased. U.S. locomotive engineers took the 
throttle on difficult stretches (and eased the long "drags" over the high iron 
on the approaches to the Naga Hills). 

The Indian block system was retained, in conformance with the policy of 
keeping Indian operating rules. The method was "not basically inefficient," but 
slow, lackadaisical operation made it appear so. "Reports of numerous inspec- 
tions of the lines invariably included reference to stops at all stations whether 
necessary or not, halts of much longer than necessary duration to obtain clear- 
ance, inefficient methods of dispatching. . . . The placing of U.S. Army Sta- 
tion Masters at many of the dispatching points along the lines and close 
supervision of all personnel resulted in considerable speed-up of traffic 
movement." 34 

With the steadily increasing forward movement of freight, the number of 
cars from other lines on the Bengal and Assam Railway inevitably increased. 
On 1 March the debit balance had been 9,600 cars. As Appleton disregarded all 
short-term considerations of balancing the movements of loaded and empty 
cars, the debit balance rose steadily. This attracted local criticism, but Appleton 
stood firm, for he expected that the cycle of return movement of empties would 
soon appear and stabilize the situation at a much higher level of traffic. In late 
April, the debit balance was 12,000 cars. 

By the end of March it was apparent that the problems of meter-gauge 
operations were well on the way to solution. Figures furnished by British 
Movements Control revealed that military traffic from Pandu and Gauhati for 

35 (1) Rpt, Appleton to Wilson, 25 Apr 44, sub: Opns by MRS of B&A Ry (meter gauge). 
Somervell File, Vol IV, CBI 1944. (2) SOS in CBI, p. 70. 
54 SOS in CBI, pp. 68, 69- 



March increased 44.6 percent over that for February, and that tonnages delivered 
to the forward areas at Manipur Road, Ledo, and Chabua were up 43 percent, 
or 34,568 tons. The meter-gauge lines now handled more traffic than the 
broad-gauge line could bring forward from Calcutta. 

The broad-gauge line from Calcutta north began as multiple track, reducing 
to double track for a distance of 124 miles, and finally to a single track for the 
last 108 miles to Parbatipur. Because of inefficient and indifferent operation, the 
line was badly congested, and loading of cars actually had to be suspended. 

Trains have been set off on sidings the length of the line definitely reducing the capacity 
by the lack of passing sidings. Crews have drawn the fires on engines, have left originating 
terminals after expiration of sufficient time to have reached final destination. Many reasons 
have been given for such a condition, such as lack of communication, insufficient motive 
power, and lack of tractive effort of power in service. It is hardly conceivable that a Broad 
Gauge system would be incapable of moving tonnage in greater volume that could be 
absorbed by a connecting Meter Gauge Railroad. But such has been the case. With proper 
supervision and operation the picture can be reversed." 

Their attention directed to this state of affairs, the British took remedial 
action. A director general of railways took over the Calcutta area, and Army 
officers with experience in railroading were stationed along the line. A pool of 
100 broad-gauge freight cars was provided at Parbatipur to handle sudden 
increases in freight from the meter-gauge lines. 36 

Operations Under Military Railway Service 

Because of the delay inherent in administrative communications the sharp 
increases in meter-gauge traffic under the Military Railway Service were not 
known in Washington until April 1944. General Arnold followed the situation 
closely. The critical supply situation in Assam, alluded to earlier in this chapter, 
moved him to tell Somervell on 17 March: "Unless the improvement of the 
rail and river lines, and the construction of the Calcutta- Assam pipe line, is 
given a more determined push, I shall be unable to furnish the service 
demanded [from the ATC]." 37 Unaware that Covell was about to report a 
34-percent jump in rail traffic for early March, Somervell at once sent him the 
gist of Arnold's strong memorandum. Meanwhile, in Delhi, General Sultan on 
21 March was taking a pessimistic view of the line-of-communications 
problem. Delhi being a great deal closer to the scene, Sultan knew of 
Appleton's progress, but even so he thought that "no matter how much they 
do the communications will never be stretched to the point where they carry 
all the stuff that the British and Americans want to send to Assam." 38 

» Rpt cited ln. WiFl 

56 Bykofsky MS. 

37 Memo, Arnold for Somervell, 17 Mar 44. Somervell File, Vol IV, CBI 1944. 

38 Memo, Sultan for Stilwell, 21 Mar 44. Item 290, Bk 3, JWS Personal File. 



The concern felt by Arnold and Somervell was speedily dissipated when 
Covell's and Appleton's reports reached their superiors. The messages came 
very close to crossing in passage. After March 1944, the Assam line of com- 
munications ceased to be a major problem; Arnold's memorandum was the last 
of its kind. 

From 1 March to 25 April 1944, or during the time when the Japanese were 
on the flood tide of their drive into India, the Military Railway Service moved 
all traffic offered to it. No restrictions on traffic were requested of higher 
authority. Even more indicative of the swift change in the condition of the 
line of communications was the fact that between 23 March and 18 April 
sixty-four troop trains were moved in addition to scheduled traffic. "In fact 
during the month of April . . . the average daily tonnage over the Assam 
LOC exceeded the British estimate of 4,400 long tons for October 1944 made 
by the Director of Transportation, GHQ (I). . . ." 39 

The War Department relieved Colonel Appleton on 27 April 1944 to send 
him to another theater. On the occasion of his departure, General Lindsell, 
Principal Administrative Officer (India), after stating that all traffic offered to 
Military Railway Service had been accepted, added: "In addition to these 
admirable technical results Col. Appleton gave us a fine example of inter-allied 
cooperation and I know that all my officers who were privileged to work with 
him will be full of regret at his leaving India." Lt. Col. Stanley Bray of the 
Railway Grand Division acted as director until Colonel Yount took command 
on 17 May 1944. There was one ugly little note. Covell reported that discipline 
among the railwaymen was not good, that there had been considerable stealing 
of military supplies. 40 

When tonnage bids were submitted by British and American authorities 
asking for priorities on rail cargo space for the month of June, they were 
accepted in full, just as March had marked the meter-gauge line's ability to 
receive all the cargo brought to it. "In July allocations for military traffic 
exceeded the target figure established at the Quebec Conference of 220,000 
short tons per month for 1 January 1946." Indeed, the Military Railway Service 
soon found the meter-gauge line had surplus capacity. 

During August, an all time high of 16,439 wagons [freight cars] was handled eastward 
over the Amingaon-Pandu Ferry which amounted to a daily average of 530 wagons. This 
figure when compared with the average of 327 wagons per day handled in March gives an 
idea of the steady increase in capacity achieved as operating experience was obtained. 
Although more American supplies were moved forward to Assam during this month than 
ever before, a part of the capacity previously allotted to the U.S. Army was turned back for 
civilian use. 

Military dispatches over the Assam LOC had grown to 6,537 long tons per day in 
September although the total available capacity was not being used. The new wagon ferry 

39 SOS in CBI, p. 74. 

40 (1) Ltr, Lindsell to Covell, 26 Apr 44; Rpt, Covell to Somervell, 22 May 44. Somervell File, 
Vol IV, CBI 1944. (2) Bykofsky MS. 



between Amingaon and Pandu and completion of many construction projects on the system 
began to be reflected in increased capacity sufficient to provide an excess of capacity of ten 
percent over actual requirements. In October the capacity for movement of military stores 
(exclusive of POL, personnel, and vehicles) exceeded demands by approximately 20 percent. 41 

Physical Improvements on the Railway 

In the fall of 1943, floods had done so much damage to the Bengal and 
Assam right of way as to cause serious concern in General Headquarters 
(India). 42 If the monsoon rains of 1944 did the same, Appleton's reforms might 
literally be washed away, and the Burma campaign endangered by supply 
failures. Thirty-odd streams cross the right of way and quickly flood during the 
monsoon. With the monsoon approaching, flood-control measures were taken 
in hand. Bridges were reinforced with heavy stone riprap to shield their piers 
from the swirling currents. 

One of the most sensitive points was the Mora Manas River bridge. It was 
necessary both to strengthen the bridge and to keep the river from flooding the 
right of way. Lt. Col. George Branch, commanding the 725 th Railway Operat- 
ing Battalion, decided to cut a diversion channel from the Mora Manas to the 
near-by Bulkhadhoba River. The 725th Battalion completed the half-mile-long 
channel in June 1944. Thanks to the several measures of flood control, for the 
first time in thirty years the monsoon did not interrupt rail traffic. In October, 
Brig. Gen. Paul F. Yount could report that Colonel Branch's channel had kept 
flood waters from doing any more than cover the tracks for a brief period of 
time, without stopping traffic. 

The right of way itself was improved and its capacity increased. New ballast 
was laid, kinky rails straightened, and low points lifted. In general, efforts were 
consistently made to bring the right of way nearer to American standards. By 
the end of 1944, the Government of India had double-tracked about 20 percent 
of the line. Of more immediate importance was the construction of a number 
of passing sidings and the enlargement of the main railway yards. Because it 
was doubtful that the double-tracking could be completed in time to benefit 
current operations, General Yount finally persuaded the Government of India 
to concentrate its attention on improving the passing sidings to speed two-way 
traffic. Though the burden of construction fell on local authority and resources, 
in many cases it was possible to furnish American earth-moving equipment to 
aid the long files of Indian laborers patiently carrying baskets of soil on their 

The problem of rolling stock was attacked by improved maintenance and 
the acquisition of U.S. equipment under lend-lease. On beginning its work, 
Military Railway Service found the rolling stock badly run down, and the 

41 SOS in CBI, pp. 75-76. 

42 For this section, the authors drew heavily upon the Bykofsky MS. 



758th Railway Shop Battalion was faced with a problem. The Bengal and 
Assam Railway, under the operating agreement, was responsible for providing 
necessary spare parts and repair facilities. Unfortunately, it had few parts at 
hand. In practice, the Military Railway Service had to requisition materiel 
through the Transportation Service. Until spares could arrive from the United 
States, there was no alternative to cannibalizing engines and cars. Despite these 
several handicaps, from 1 March to the end of 1944, the 758th Battalion repaired 
47,044 cars. 

A great portion of the Bengal and Assam's motive power was of U.S. origin 
via lend-lease. When the Military Railway Service took over operations on 
1 March 1944, 167 of the 396 locomotives on hand were War Department 
engines. By 30 September, the number was 203, over half the total on hand, and 
by the end of the year the figure was 238. With the engines came a steady flow 
of American meter-gauge freight cars, adding up to some 6,500 by the end of 
1944. These were the commodious American boxcars, with twice the capacity 
of the little four-wheel cars used by the Bengal and Assam. Military Railway 
Service estimated that the capacity of the lend-lease rolling stock was, by the 
end of 1944, equal to that of the railway's own equipment. 

Operation of the lend-lease boxcars was made more difficult because Indian 
shops, in erecting them, omitted brakes and vacuum hose. This often caused 
collisions and the breaking-in-two of trains. By the end of 1944 the shopmen 
were able to equip a sizable number with proper brakes. The end of this prob- 
lem was foreshadowed when, in October, the Railway Board of the Govern- 
ment of India promised that all cars assembled by them would be properly 
equipped, and that they would provide 600 brake sets per month for cars 
already in operation. 

A British Appraisal 

The quick and dramatic improvement in the operations of the Bengal and 
Assam Railway was perhaps the outstanding example of successful application 
of American technology to the logistical problems of war in the China, Burma 
and India Theater. An editorial in the Indian press stated: 


Railway communications in Eastern Bengal and Assam were never good. Difficulties of 
terrain and climate had been taken as excuse for sloth. Defence needs were ignored in India's 
north-east corner in odd contrast to its north-west one where there was lavish construction 
of strategic railways and roads. When the Japanese threat developed in 1942, there ensued 
sudden sharp deterioration in the efficiency of railways near the Eastern Frontier, particularly 
in Assam, owing to the wholly unwonted and unforeseen increase of traffic. External pressure 
and internal administrative stress on the communications system coming about just at the 
time when the need for quick movement was imperative were moreover soon aggravated by 
India's political disorders. The situation became insufferable; journeys took days; even deaths 



en route were not infrequent. This state of affairs prevailed in the latter part of 1942 and 
continued, little modified, throughout 1943; drastic action was obviously necessary. 

This came at last in the beginning of 1944 when sections of the Bengal and Assam 
Railway were turned over to American operational control. U.S. enterprise and hustle have 
succeeded in this instance, most remarkable results being achieved which (according to 
common report) British and Indian experts on the basis of much local experience had stoutly 
predicted were impossible. What has been done must be most gratifying to Maj-Gen Covell 
. . . completion of a job of extreme difficulty which must be regarded as a major contribu- 
tion to the Allied victories in Burma. Both the Government of India's Railway Member and 
the Manager of the B and A Railway have praised the work of the U.S. railway battalions, 
first under Col. J. A. Appleton and later under Brig-Gen P. F. Yount. Thousands of Indian 
railwaymen cooperated with the Americans and though difficulties were inevitable, they 
were surmounted through a "remarkable degree of tolerance on both sides." 

As the Railway Member said recently, the Americans "got their coats and shirts off to 
the task" in Assam. That is their way and perhaps was the only way to have completed at a 
critical juncture what the official statement . . . describes as the lifeline for the troops in 
Burma. 43 

In his final report to the CCS, Mountbatten contented himself with remark- 
ing: "The immediate increase in deliveries to Assam over the railway was not 
as great as Lieut.-General Somervell had hoped, but during the latter part of 
1944 and the early part of 1945 the lift rose very rapidly; and the introduction 
of military control was unquestionably justified by the results obtained." 44 

Attempts To Use Indian River Transport 

Since the Assam line of communications lay in the valley of the Brahma- 
putra, one of the world's great rivers, the use of inland water transport to 
supplement the rail line seemed an obvious approach, and a great deal of effort 
was expended in trying to make the river a more efficient artery. Barge lines 
under civilian operation were bringing up small quantities of supplies to the 
various river ports— many of them little more than landing stages. After the 
Trident Conference (Washington, May 1943) decided not to attempt to 
retake all Burma and reopen the line of communications from Rangoon north, 
SOS planners suggested that barge equipment originally intended for use on 
the Irrawaddy River be placed on the Brahmaputra. The project was approved 
at the Quadrant Conference, Quebec, August 1943, with the target date of 
1 April 1944. 

Over the six months' period September 1943-March 1944, experiments in 
the United States and in the CBI Theater revealed that the tugs and barges 
procured were not very well suited to their task. The question became one of 
using them to best advantage somewhere in India. Meanwhile, the sharp 
improvement in operation of the meter-gauge rail lines made the issue less 

4 ^ The editorial (Statesman, May 27, 1945) appears as an inclosure of the Covell Report. 
44 Mountbatten Report, Pt. A, par. 46. 



By July 1944 Transportation Service concluded that the equipment and men 
at hand, many of the latter highly skilled in barge-line operation, could best be 
used for lighter and towing service around Calcutta, ferry and terminal service 
on the Brahmaputra River, and moving POL (gasoline [petrol], oils, and 
lubricants) and dry stores to airfields in East Bengal. 45 Initial hauling of POL 
by barge from Goalundo, on the broad-gauge rail line, and Dacca in East 
Bengal was begun on 17 August. Hauling of dry stores from Khulna to Dacca 
began soon after. In September, the peak month of fall 1944, 1,934 long tons of 
stores and 601,474 imperial gallons of POL were carried from Goalundo to 

The civil barge lines on the Brahmaputra River came under the jurisdiction 
of the Assam Line of Communications Panel, which integrated their operation 
with that of the railway. The principle followed was that of cutting long barge 
hauls from Calcutta to Assam to the minimum, using instead rail from Calcutta 
to Sirajganj Ghat and Dhubri, ports on the Brahmaputra, thus cutting down 
the barges' turnaround time. 46 

Pipelines in India 

There were two major U.S. pipeline projects in India, one supplying the 
B-29 bases around Calcutta, 47 and one from the Budge-Budge oil terminal of 
the Burma-Shell Co., about sixteen miles below Calcutta, to a tank farm at 
Tinsukia in upper Assam. The latter line was an integral part of the Assam 
line of communications. 

For construction purposes, the line was divided into four areas, each under 
an area engineer. The American portion of the construction work was done by 
the 700th, 708th, 709th, 776th, and 777th Engineer Petroleum Distribution 
Companies. Col. William C Kinsolving was in charge of the project for Engi- 
neer Construction District No. 12. Because of the problems of labor recruiting, 
land use, housing, and local procurement, a British garrison engineer and five 
assistant garrison engineers were attached, and the project, as usual in India, 
became a combined one. Approximately Rupees 4,000,000 worth of work was 
contributed by agencies of the Government of India. 

Following a period of reconnaissance and design on the 752-mile line, work 
began 25 February 1944. Because the Assam line of communications was badly 
congested at this time, deliveries of material were slow, and the work did not 
really get under way until April. The material itself was a headache to the 
construction crews. Pipe came in tons before couplings were delivered. Inex- 
perienced port and freight personnel lost and damaged a good many pieces of 
equipment. Very few standard spare parts ever arrived. To the pipe-laying crews, 

45 (1) Bykofsky MS, pp. 115-16. (2) SOS in CBI, pp. 96-98. 

46 ( 1 ) Bvlmftk v MS, p. 1 19. (2) SOS in CBI, p. 99. 

47 Se j Ch. IHJ above. 



sweating away in the Indian heat, it must have seemed as though a malign 
goblin was deliberately hiding the essential small parts, fittings, and screws 
without which the imposing devices of modern industry will not work. But 
the work went on, with the Indian artisan, ingenuity, and field expedients doing 
their best to fill the gaps. 

The first 300 miles of pipe north from Budge-Budge were on the Bengal 
and Assam's track just outside the rails. The water table in that area is so high 
during the monsoon that ditches were not practicable, and the local road net 
was so poor that there was no other means of access. The staff of the railway 
could not have been more co-operative. The Indian railwaymen scheduled work 
trains so carefully that pipe was always in place along the tracks before the 
pipeline crews arrived. Rivers were crossed by pulling prewelded sections of 
line across them by tractor. 

The fragility of the invasion-weight pipe used and the heavily populated 
countryside offered serious problems of construction and operation. The West- 
erner, and particularly the American, really does not comprehend the phrase 
heavily populated until he has seen the Indian countryside, with village following 
on village, sometimes only a mile or two apart, and the space between filled 
with little plots like suburban gardens. Where the safety of the villagers seemed 
to require it, the pipe was buried, and because of the dense population this 
meant burying long stretches of pipe. This in turn created two more problems: 
finding the labor to bury the pipe, and the rapid corrosion of the thin-walled 
pipe once laid. 

Coolies were hired by the garrison engineer and his assistants through 
piecework contractors. The local contractors had never been faced with a job 
of such magnitude, particularly since it involved keeping a labor force moving 
forward. The laborers could not go far from their villages, for no surplus of 
food or housing would be available in the countryside. And the contractors' 
personnel policies, if they can be so dignified, were blends of inefficiency and 
time-honored skulduggery. When the Calcutta area was left behind, it proved 
best to put the laborers on the payroll directly, give them an identity token, pay 
them daily, and supervise them with American noncommissioned officers. Con- 
struction of three pipeline plows by District No. 12 shops proved very success- 
ful. They equaled 1,000 laborers for digging ditches in which to lay the pipe. 

The thin-walled pipe was not only an operating problem but a hazard. 
Experience with 160 miles of it in West Bengal, in the B-29 supply system, 
suggested that the loss would rise steadily as the pipe remained in the ground 
and that it would not require many months to reach significant proportions in 
a long line. Hundred-octane gasoline seeping ever farther through the soil of a 
populated area is a deadly hazard. 

On 26 June 1944 a leak was found where the pipe crossed the Hooghly 
River near the village of Ulabaria. The crews began to repair the line, follow- 
ing normal operating and safety procedures. Water was pumped into the line 



and all gasoline seemed to be clear. The odor of gasoline in the vicinity did not 
seem heavy; moreover, a soldier was posted in the street warning the villagers 
not to start any open fire. Everything appeared in order and the crew prepared 
to test the pipe with compressed air for leakage. As the sergeant in charge 
walked toward the open end of the pipe, there was a vapor explosion. There 
was no subsequent vapor fire but the villagers' thatched huts promptly ignited. 
In the resulting holocaust, seventy-one people were burned to death, and about 
twenty buildings destroyed. After the Ulabaria tragedy it became customary to 
use standard-weight pipe in dangerous areas. 

The pipeline was complete from Budge-Budge to Tinsukia on 8 July and 
testing began on the 13th. The entire system was in operation on 14 August, 
though, as noted above, sections of it began operating earlier to relieve pressure 
on the railway. 

The same month, August, that saw completion of the first pipeline from 
the Calcutta area to upper Assam also saw War Department approval of a 6-inch 
line from the port of Chittagong to Tinsukia. Such a route would be shorter, 
have no major river crossings, be flood-free during the monsoon, and give the 
advantage of dispersal and extra facilities. Its capacity was estimated at 36,000 
long tons a month. Construction began on 16 October with emphasis on 
completing the section from Chittagong to Tilagaon to connect with the East 
Bengal fields of the ATC. 48 

Supply Problems in India 

After General Covell's reorganization of SOS, and until late 1944, there 
were no major structural changes in the SOS. A War Department change in 
the authorized supply levels was successfully complied with, thanks to the 
improved Assam line of communications, and the supply of reciprocal aid from 
India gave signs of approaching the maximum. It was a period of steady 
improvement in efficiency of operations, in stocks on hand, and in ability to 
carry out the SOS mission. 

General Covell's reorganization plan called for Advance Section No. 2 to 
become an Intermediate Section and expand northward to take over the Ledo 
base. Covell had postponed implementing this project lest it interfere with 
General Pick's building the Ledo Road. When in May 1944 the issue was raised 
anew, Pick opposed the change on the ground that his organization, fully 
extended in support of current operations, would suffer if such a major ampu- 
tation was performed on it. His views prevailed, and though an order directing 
the change was published 23 August 1944, shortly after Myitkyina fell, the 
actual changes were not made until considerably later. 49 

48 The above is based on: (1) SOS in CBI, Appendix 12, entitled Construction Service, which 
contains a Report on Pipeline Program Carried Out by Engineer District No. 12. (2) SOS in CBI, 
pp. 115-16. 

49 SOS in CBI, pp. 335-56. 



In January 1944 the War Department took on itself the responsibility of 
determining supply levels. Immediately thereafter, it cut the maximum level of 
supply in CBI for Class I and Class III supplies from 180 to 120 days, of which 
30 were operating and 90, reserve. Theater headquarters protested that the 
30-day operating level was insufficient because of the length of time it took to 
move supplies from Calcutta to the front. Fortunately, the sharp increase in 
deliveries over the Assam line of communications that ensued in March 1944 
relieved concern on this point. Indeed, such was the over-all improvement in 
the theater supply situation that in compliance with a War Department direc- 
tive ordering periodic review of supply levels to reduce them where possible, 
SOS recommended that authorized levels for Class II and IV supplies also be 
cut from 180 to 120 days. 50 

New Agreements on Local Procurement 

Because the U.S. forces in CBI had been directed to make the maximum use 
of local resources, and because the Government of India was most co-operative 
in supplying them as reciprocal aid, local procurement was a major interest 
of SOS. 

In the summer of 1942, SOS, CBI Theater headquarters, and the Govern- 
ment of India reached an agreement which worked so well that in fall 1942, 
General Wheeler, then commanding SOS, could report to General Stilwell that 
the U.S. forces were living off the land. 51 

A group of new agreements with the Government of India in spring 1943, 
following the visit of a lend-lease mission from the United States, extended and 
speeded the reciprocal aid process. The sometimes-debated question of the 
scale on which India would issue supplies to U.S. troops was settled in June 
1943 by India's agreement to issue supplies without reference to the British 
scale. The SOS agreed not to ask for supplies in excess of the normal American 
scale, and further, that save in emergency it would not requisition imported 
stores from India, except for oil products, which had a special status. India 
speeded the process of supply by permitting SOS to requisition rations from 
the nearest Royal Indian Army Service Corps depots. SOS in turn promised not 
to buy imported stores on the open market in excess of Rupees 10,000 without 
permission from the Government of India. To keep this latter agreement by 
SOS from hampering the American supply position in emergencies, the Gov- 
ernment of India soon after agreed that the SOS could apply to the nearest 
Indian Army depot for imported or domestic stores other than foodstuffs in 
amounts up to Rupees 10,000. 5 2 

In February 1944, the process of liberalization reached a probable limit when 
U.S. forces in the field were enabled to requisition certain items common to 

50 SOS in CBI, p. 344. | . 

51 Stilwell 's Mission to ChinalCh. VI. I 
» SOS in CBI, pp. 144-47. 1 1 



Indian and U.S. Army use directly from the local Indian Army depot without 
securing approval from SOS headquarters or General Headquarters (India). In 
effect, this gave U.S. organizations the same status as that of neighboring 
Indian units in drawing common-user items from Indian stocks. 

During 1943 and 1944, reciprocal aid from India steadily increased until it 
finally began to seem that the upper limit was being approached even though 
various U.S. agencies took steps to improve Indian productivity. Indians were 
trained in the United States. Teams of U.S. technicians toured Indian factories 
to teach improved production methods. Raw materials, such as steel and 
alkaline-reclaimed rubber, and capital goods, such as rolling stock and mining 
equipment, were furnished where advisable and possible. 53 

Indian civilian employees, ranging from Anglo-Indian secretaries in head- 
quarters installations to Indian technicians to unskilled Indian laborers were 
an essential and invaluable reinforcement. Since the American effort in India 
was primarily air and supply, employment of these Indian civilians meant that 
fewer American troops were needed. These Indians were paid by the Govern- 
ment of India under reciprocal aid. By 30 September 1944, the SOS and AAF's 
Air Service Command employed approximately 79,000, of whom 21,000 were 
skilled and semiskilled, and 58,000 unskilled. Many of the skilled people 
learned their trades while working for the U.S. Army. Training and supervising 
them were added to the responsibilities of many U.S. Army enlisted men, so 
that the advantage was not without its indirect cost. 54 

In 1944, SOS could only estimate the dollar value of the reciprocal aid 
extended by the Government of India. That government claimed its adminis- 
trative resources would not permit furnishing an appraisal of the dollar value 
of reciprocal aid. The SOS therefore was obliged to set up its own system for 
appraising reciprocal aid received so that it might have data both for current 
operational purposes and for the anticipated postwar settlement of accounts. 55 
The SOS estimates for the period from January 1943 through October 1944 are 
given below: 

1943 1944 

January-February $5,710,370 $35,662,824 

March-April 6,413,377 46,042,316 

May-June 9,049,295 52,482,591 

July-August 13,113,736 56,203,485 

September-October 16,881,941 57,050,847 

November-December 27,469,293 

In February 1944, the Finance Member of the Government of India, by an 
address to India's Legislative Assembly, gave clear warning that the Govern- 
ment of India was about to combat inflationary tendencies in the Indian 

» SOS in CBI, pp. 149-50. 

54 G-4 Per Rpt, Hq USAF CBI, quar ter ending 30 Sep 44, pars. 3i, 9g. OCMH. 

55 St if well's Mission to China fp^ zvy.f 



economy by cutting back military production. He hinted that if India was to 
furnish more support to operations in Southeast Asia, then her commitments 
for support of Middle Eastern and Mediterranean operations would have to be 
reduced. At the same time, official circles of the Government of India com- 
mented that 1944 supply would not be greatly affected by the new policy but 
that major cutbacks in military production could be expected in 1945. 

Soon after, production of woolen cloth for military use stopped. All output 
was reserved for civilians, and by the end of 1944 many types of cotton goods 
were reserved for civilian use. Trying though this may have been for American 
supply personnel, nevertheless, the Indian point of view should not be over- 
looked. The cash income of that statistical fable, the average Indian, was about 
$20.00 a year (Rupees 60). From that sum, he had to buy such cash items as 
cooking oil and salt, and supplement his meager diet. Very likely the peasant 
family wardrobe included one set of garments per person. Thus, there was no 
slack to be taken up, and the peasant counted on being able to replace clothing 
as it wore out. If no cloth was available at the bazaar, the little spinning wheels 
of the villages could hardly make up the difference. Sullenness in the villages, 
whence comes the Indian soldier, could soon imperil the whole Indian war 
effort, and the Government of India had to realize that there was a point 
beyond which the patient farmer could not be overloaded. 

In the light of the warnings referred to above, SOS could not have been 
surprised to learn in July that 183 of 336 items previously available at Indian 
depots on blanket sanction would be unavailable in differing percentages, some 
as high as 100 percent. Similarly, the Government of India estimated that 246 
items of 648 would be unavailable in 1945. Beginning 1 August 1944, the 
quantity of dairy products, meat, sugar, and fish that India would supply was 
cut back to the May 1944 level. Indian authorities said that they wanted to 
cancel supplying steam-laundry facilities, office supplies, and printing. General 
Covell personally interceded to have that action rescinded. 56 

In addition to applying pressure at the policy level, the Government of 
India took unilateral action to cut the flow of reciprocal aid. General Covell 

Commitments in India had come to have none of the firm connotation associated with 
them in America. Promised delivery dates were ignored or quantities reduced without 
advance notice, often months after promised delivery dates. Many commitments for supplies 
were made but actual delivery was frequently late, often in quantities less than required, 
making it difficult for SOS to plan future supply. Frequently emergency shipments from the 
United States became necessary because last minute cancellations by Indian agencies would 
not permit requisitions on the U.S. to be processed normally. 57 

The Indian attitude affected Sino- American relations because the United States 
understood that His Majesty's Government in the United Kingdom would 

56 SOS in CBI, pp. 153-55. 

57 Covell Report, pp. 31-32. Substantially the same passage is in SOS in CBI, p. 156. 



supply the Chinese forces based on India through British lend-lease to China. 
But, in "many cases" this had not been done, and approximately 33 percent of 
the total value of supplies to the Chinese Army in India came from U.S. sources, 
$9,000,000.00 worth. 58 

The situation was complex. Neither SOS nor the Government of India 
desired to go beyond a certain point, and so there was room for compromise, 
adjustment, and negotiation. SOS took full advantage of the diplomatic oppor- 
tunities open to it, and succeeded in actually raising the dollar volume of 
reciprocal aid as between the May-June 1944 period and that of September- 
October. Examination of physical volume suggests that oil products increased, 
subsistence held steady, and textile items dropped. Despite the present success 
of its diplomacy, SOS felt concerned about the future. 59 

Housekeeping Problems 

Difficulties in the movement of supplies to the forward areas in Assam, 
further complicated by inadequate warehouse space, made stock control in 
Assam very difficult in 1942 and 1943. Since the base depots automatically for- 
warded balanced stocks, the lack of proper stock control ultimately led to an 
actual overstocking of certain items, particularly quartermaster supplies. 
Because the quartermaster had to provide food for an extraordinary collection 
of races and sects (Hindu, Moslem, Burmese, Chinese, hillman, British, and 
American) each with different dietary requirements or regulations, it was nec- 
essary to keep an extremely varied inventory, with all the administrative prob- 
lems involved. Beginning in September 1943, a stock record system for 
quartermaster items was set up. It did not work well because the warehouses 
were not well organized. 

The War Department in September 1943 instituted a system of automatic 
supply based on reports from overseas theaters, which in turn had to be an 
accurate reflection of stocks on hand. This required changes in CBI procedures. 
Stock control and inventory teams were sent out from the United States to 
place the system in operation, a task of several months. 

Another problem was that of securing an adequate supply of spare parts. 
War Department procedure required spare parts to be requisitioned on a 
formula basis. Unfortunately, the formula did not fit operating conditions in 
Assam and Burma. Abrasive dust in the air wore down engine parts. Fungus 
and rust attacked delicate instruments. Not until requisitions could be based on 
usage factors as determined by actual experience in the theater, was it possible 
to obtain adequate stocks of spare parts. 60 

58 SOS in CBI, p. 156. 

59 (1) Covell Report. (2) SOS in CBI, tables facing p. 167. 

60 SOS in CBI, pp. 334-400. 



Chinese Lend-Lease 

For SOS, Chinese lend-lease was a housekeeping problem, for its responsi- 
bilities were executive, not policy-making. Housekeeping in the most literal 
sense was badly needed for Chinese lend-lease, for as late as the fall of 1943, 
4,200 tons of Chinese lend-lease were still to be seen lying in the Assam rice 
paddies where the agents of China Defense Supplies, Inc., had dumped them 
in the spring of 1942. They had lain there through three seasons of monsoon 
rains and scorching heat. For example, there were in India 261,000,000 rounds 
of 7.92-mm. ammunition for China. One hundred million of them were in 
Assam. By February 1944, it was estimated that heat and moisture had oxidized 
80,000,000 rounds. SOS asked General Headquarters (India) to renovate the 
whole store in Indian arsenals. 61 Not only was lend-lease not properly stored, 
but the number and nature of items on hand were unknown. 62 

The process of honoring earlier lend-lease commitments to the Chinese 
created what SOS knew as the stockpile problem. As the shipping situation on 
the high seas eased, lend-lease for China began to move to India, there to wait 
shipment to China. Since by Presidential decision, the bulk of Hump tonnage 
was to go to the Fourteenth Air Force, the prospects of moving these stockpiles 
to China were remote. Moreover, the International Division of Army Service 
Forces had been shipping materiel as it was available, against General StilwelPs 
requisitions for sixty Chinese divisions. 63 This meant that the stockpiles were 
out of balance, with critical items in short supply, and warehouse facilities 
filled with less vital needs. 

The ever-growing bulk of the stockpiles, evenly divided between military 
and civilian goods, was a handicap in procuring items of current necessity 
because the Munitions Assignments Board in Washington (which allocated 
lend-lease in accord with directives from the Combined Chiefs of Staff) tended 
to see the bulk of the piles rather than their composition. Moreover, SOS 
officers could see little logic in their having to requisition items for the U.S. 
Army from the United States, with all the delay involved, when identical pieces 
of lend-lease equipment were on hand in India with their prospects of delivery 
to China very remote. 64 

On 31 December 1943 Col. William S. Gaud, Jr., took up his duties in 
Chungking as representative of the War Department in "all matters pertaining 
to the assignment of military supplies and equipment to the Chinese Govern- 
ment." This arrangement made it possible for the Chinese Government to deal 
directly with the War Department on the sensitive issue of lend-lease and 

61 SOS in CBI, pp. 465-67. 

62 Memo, Col William S. Gaud, Jr., WD Mil Aid Representative to China, for Director of 
Materiel, Hq ASF, 10 Dec 43, sub: Lend-Lease to China, in Gaud Rpt 2, 4-10 Dec 43. (Here- 
after Gaud Report .) AG (ASF ID) 319.1, A46-299. 

63 (1) SOS in CBI, p. 470. (2) ForB^JifCussioji^f the origins of the sixty division program, 
see Stilwelh Mission to China, Chaptersj VIl|anc |VIII. | 

64 SOS in CBI, p. 471. 1 1 1 1 



freed CBI Theater headquarters of the unpleasant duty of commenting on 
Chinese lend-lease requisitions. 

Colonel Gaud's survey of the physical aspects of the lend-lease situation in 
Assam disclosed some of the arguments against pouring lend-lease into a 
command without the physical resources to handle it, and of shipping some of 
it under the auspices of an organization, in this case, China Defense Supplies, 
Inc., which did not function with administrative efficiency. Colonel Gaud 
found that no one knew exactly how many tons of Chinese lend-lease were in 
India nor exactly what was on hand. China Defense Supplies, Inc., did not al- 
ways give adequate identification of what it shipped to India, nor did it always 
furnish manifests or shipping lists. Once goods arrived in India, they were 
handled by SOS, which was perennially short of American personnel and de- 
pendent on the Government of India for warehouse construction. Goods were 
manhandled from ship to train to shed by coolies ignorant of what fragile or this 
side up might mean. Once in the sheds, supplies were simply piled there, with 
no attempt at physical inventory. Records showed only "CDS supplies" or 
"spare parts." CDS storage areas in Assam were not fenced in, and were 
guarded only during daylight hours by a few unarmed Chinese civilians. Colo- 
nel Gaud believed the basic cause of wastage was a shortage of American en- 
listed personnel. He found SOS fully aware of the problem, but completely 
handicapped by the simple lack of men to do the work. Therefore, a situation 
had arisen in which Headquarters, SOS, and the subordinate sections gave to- 
tals of lend-lease on hand that differed by as much as 50 percent. 65 

Since SOS was fully appreciative of the situation, the corrective measures 
and proper storekeeping could be applied as reinforcements came to the thea- 
ter in accord with the Quadrant decisions to provide an adequate logistical 
foundation for the Allied effort in Asia. Thus, SOS personnel strength in- 
creased fro m 31,074 on 30 November 1943 to 42,574 on 29 February 1944. (See 
Table 4. so, on 28 February, Colonel Gaud could report to the War De- 

partment that the situation in Assam was "markedly different" from what it 
had been in December. 66 With the inventory problem being solved, Gaud 
thought that the warehouse construction problem was the principal one to be 
faced in handling Chinese lend-lease. 

In spring 1944 the inventorying and proper storage of Chinese lend-lease 
made diversions from it to fill U.S. supply needs administratively simple even 
if diplomatically complex. Ever since the opening of the first U.S. base at 
Karachi, the Chinese had been willing to permit diversion from their lend- 
lease stocks to the SOS. With the decision of the War Department in the 
spring of 1942 that General Stilwell should decide when and where title would 
pass to the Chinese, there were no legal barriers to repossession of War De- 
partment-procured Chinese lend-lease that arrived in India after the summer 

65 Gaud Reports 1 and 2. 

66 Gaud Report 7. 



of 1942. However, Stilwell knew what the effect of a policy of wholesale diver- 
sion would be on the Chinese. Therefore, until May 1944, SOS was ordered to 
follow a policy of securing the consent of Chinese authorities before diverting 
any lend-lease item to American use. 67 

In April, the Munitions Assignments Board in Washington ordered CBI 
Theater to repossess all War Department-procured lend-lease vehicles, tires, 
and spare parts in India beyond those needed for the Chinese Army in India, 
for the U.S.-sponsored Chinese divisions along the Sal ween, and for Ramgarh 
Training Center. The order specifically exempted nonstandard trucks and parts 
ordered by the Chinese for use in China. The Munitions Assignments Board 
further authorized the theater to divert any War Department-procured lend- 
lease that might be used to advantage in the war effort. 

Gen. Ho Ying-chin, Chief of Staff of the Chinese Army, was told on 3 May 
that the Chinese would be informed of all diversions. The Chinese would be 
allowed to requisition replacements, but the Munitions Assignments Board 
reserved the right to pass on all such. The reason for the policy, General Ho 
learned, was to make effective use of all supplies that could not be transported 
to China. 68 

The Chinese asked that Headquarters, CBI Theater, promise to replace all 
diverted lend-lease, and in Washington, T. V. Soong protested strongly. The 
reply to Dr. Soong said: 

There existing in the United States a growing aversion to idle stockpiles, especially in 
view of spoilage and wastage, and an ever-increasing desire that all materials be used to the 
best possible advantage in the war effort. . . . The State Department and Foreign Economic 
Administration are in complete agreement that under the Master Agreements title to all 
lend-lease material of whatsoever nature and to whatsoever country transferred remains with 
the United States Government. . . . 

The United States Government retains full authority for the diversion of such China 
Lend-Lease supplies as are incapable of serving the end use for which they were originally 
provided. . . , 69 

In compliance with the Munitions Assignments Board directive, SOS was 
authorized to divert War Department-procured materiel to U.S. troops or issue 
such to U.S.-sponsored Chinese divisions, without consulting the Chinese. 
Initially, diversions under this policy were small, and until 1 July 1944 totaled 
only $5,000,000 for medical, ordnance, signal, and engineer supplies. 70 

Medical Problems in the Rear Area 

The medical problems of operations in India and Burma were potentially 
grave. They reflected an environment that was either inherently unhealthful, 
as in Burma, or one that had, as in India, become unhealthful through the un- 

67 SOS in CBI, p. 475. 

68 Ltr, Ferris to Ho, 3 May 44. Gaud Report 12, Tab A. 

69 Ltr, Walter W. Fowler, Special Representative, to T. V. Soong, 5 Jun 44. Gaud Report 14, 
Table B. 

70 SOS in CBI, pp. 475-76. 



sanitary practices of the inhabitants. In India, fevers and alimentary disorders 
were endemic. In Burma, the insect-borne diseases such as malaria and scrub 
typhus were everywhere, while the difficulties of maintaining sanitation on 
campaign in a hot, moist climate spread alimentary diseases. Fortunately, the 
American medical authorities in the theater, who functioned as part of the 
SOS, were able to draw on local experience and resources. Hospital space, sup- 
plies, and counsel were freely given by British and Indian medical organiza- 

In October 1943 the SOS medical organization was deployed from Karachi 
to Ledo. There were probably close to 10,000 beds available for patients. The 
exact situation is obscure because hospitals operated more beds than their 
T/O's provided^ about 75 percent more on the average. At Ledo the 20th 
General Hospital and the 14th, 48th, and 73d Evacuation Hospitals had 3,250 
T/O beds. In central India, at Agra and Delhi, were the 97th and 100th Station 
Hospitals, with 100 T/O beds each. At Karachi was the 181st General Hos- 
pital, with 1,000 T/O beds. In China, whose support was the aim of all Amer- 
ican effort, was the 95th Station Hospital, 100 T/O beds, at Kunming. 

Working with the 20th General and 14th, 48th, and 73d Evacuation Hos- 
pitals in the Ledo area was the 151st Medical Battalion. The function of the 
151st was to link the combat area with base medical units. Initially, it estab- 
lished a clearing station and a chain of collecting stations, which were even- 
tually outmoded by the growing practice of air evacuation. The most signifi- 
cant work of the medical units from Ledo forward was medical care for the 
Chinese Army in India. 71 

The dispersal of American resources among the airfields of Assam and the 
United Provinces (Agra and Oudh), along the line of communications from 
Calcutta north, and in the port cities— a dispersal resulting in a great number 
of small extemporized detachments— forced an equal dispersal of medical care. 
Sixteen dispensaries were being operated in Calcutta and Karachi. Pipeline 
construction crews, railway personnel, and Signal Corps crews were followed 
by small provisional medical units. In early 1944 unauthorized provisional hos- 
pitals (not activated as T/O units) were operated at Kweilin, China, Kurmi- 
tola, Bengal, and Kanchrapara and Camp Angus, Bengal. Three more un- 
authorized hospitals were under construction. Personnel for these many ad hoc 
installations came from the several hospitals which were correspondingly 
strained when they needed facilities to care for the American and Chinese 
troops in their vicinity. The necessity of providing hospitalization for the Chi- 
nese in medical installations whose facilities had been calculated by the num- 
ber of U.S. troops on hand was a further severe strain. Hospitals, perhaps 50 
percent of whose personnel were on detached service, had to cope with a 
patient load far above their normal capacity. 72 

71 SOS in CBI, App. 16, Medical Section, Sec. I, 1942-44, pp. 31, 47. 

72 App. cited n. 71, Sec. I, pp. 52-53. 



Theater authorities believed that the situation was not understood in Wash- 
ington, for medical statistical reports reflected only the hospitalization provided 
for U.S. personnel. In June 1944, the Tables of Organization for hospitals in 
the theater provided 8,800 beds. Of these T/O beds, 7,787 were occupied, so 
that, formally, the picture was comforting. Actually, there were 18,635 beds 
available, or 212 percent of the T/O capacity. Of these, 12,530 were currently 
in use, 7,130 by Americans and 5,400 by Chinese. Since medical personnel had 
been allotted for only 8,800 beds, the strain was obvious. Energetic representa- 
tions were made to the War Department, and the CBI Theater surgeon, Col. 
Robert P. Williams, presented them in person. 

While CBI Theater waited for these requests to bear fruit, a program of re- 
organization and redeployment was begun within the theater. For example, the 
1,000-bed hospital at Karachi, a port declining in importance as Calcutta grew, 
was reorganized as a 500-bed unit. The personnel and facilities thus released 
were used to activate the 371st Station Hospital which, in turn, went to Ram- 
garh, there to operate a 750-bed hospital. Other static hospitals were enlarged. 

The Surgeon General's Office agreed to send two 1,000-bed general hospi- 
tals, one400-bed field hospital, and two 100-bed station hospitals to the thea- 
ter, plus medical personnel to permit expanding units in CBI and activating 
others. In the fall of 1944, the enlargement of six hospitals and the arrival of 
others from the United States greatly eased the personnel shortage. 73 

The 1943 hospital admissions suggest the problems faced in India. There 
were 8,136 admissions for malaria; 6,744 for alimentary disorders; 2,637 for 
venereal disease, and 1,150 for dengue fever. 74 The first malaria control and 
survey units arrived in early 1943. A program of mosquito control was em- 
barked on, but results were not satisfactory. In this connection, it must be re- 
corded that DDT was not available in quantity in CBI until the end of 1944, 
nor was atebrin suppressive therapy used in India proper. Only in the combat 
zone (roughly, forward of Ledo) was atebrin used, and it was late 1943 before 
the use of it was made general. 75 

Because theater medical authorities believed their attempts to control mos- 
quitoes were effective, they were disappointed at the results obtained in 1944. 
The CBI malaria rate in 1943 was 206 per thousand per year (250 in the Cal- 
cutta area). 76 In 1944 it dropped to 167 per thousand per year. This did not 
seem to compare well with experience in the Southwest Pacific. The medical 
men were therefore inclined to believe that there had been a failure in dis- 
cipline and personal protective measures, by implication, a failure in command 
and morale. In the light of this estimate, a campaign was undertaken in late 

7> App. cite J n. 711 Sec. II, 1 Jul-24 Oct 44, pp. 13-18. 
14 App. citedLjiJ Sec. I, p. 46. 

75 (1) Ston e MS. II 38 1. (2) App. cited n. 71, Sec. II, pp. 37-38. 

76 App. citec |n. 19(1)1 Sec. Ilf. 



1944 to indoctrinate all echelons of the theater in command responsibility, 
malaria discipline, suppressive atebrin, and the use of DDT. 77 

The number of alimentary disorders admitted to hospitals was not a fair 
index of the seriousness of that problem. In the hot season of April to August 
1944, the time of the year when disease rates always rose sharply, there were 
16,562 cases serious enough to be marked either hospital or quarters. The doc- 
tors estimated that three times that number of men were affected, but able to 
stay on duty at whatever cost in discomfort and loss of efficiency. Nor can the 
factor of lowered vitality and predisposition to other diseases be ignored. All 
in all, perhaps 200,000 man-days were lost from this cause in 1944, a serious 
matter in an undermanned theater. "It is believed that this high rate was caused 
by poor mess sanitation, employment of native food handlers, inadequate re- 
frigeration, inadequate water heating facilities, failure to control flies, insanitary 
disposal of human excreta, contaminated water, eating at civilian restaurants, 
and a defeatist belief among officers and men that such diseases were inescap- 
able in this Theater and that the sanitary problem was unconquerable." 78 

That alimentary disorders were endemic in India was no secret and the first 
American arrivals sought to meet the problem by radical measures. The base 
section veterinarian at Karachi in 1942 surveyed the local abattoir, ice plant, 
and dairy, and condemned them all. He recommended that the SOS build and 
operate its own food-processing installations, a step that became standard at all 
major U.S. troop concentrations. As American personnel were deployed across 
India the burden on the Veterinary Corps grew apace. "In addition to food in- 
spection, they supervised the operation of abattoirs; supervised the production 
of dairy products; operated small farms; scoured the country for beef animals; 
assisted in vaccination of local herds against rinderpest and anthrax; and as- 
sisted local Quartermasters in closing contracts for food, and in the overhaul- 
ing of subsistence stores." 79 The Sanitary Corps took analogous precautions 
with the water supply. 

In addition to providing pure food and water, the medical authorities 
sought to prevent later contamination of foodstuffs. The personnel shortage, 
which made it hard to obtain kitchen help and waiters, made it necessary to 
employ Indian civilians. So far as possible, they were assigned duties which 
did not involve food handling. Attempts were made to provide adequate sup- 
plies of hot water, refrigeration, screening, garbage disposal facilities, and DDT 

Unfortunately, these efforts did not bring the results desired. Admission 
rates per thousand per annum for diarrhea and dysentery were, until August 
1944, substantially higher than in 1943. Beginning in May 1944. the malaria 
admissions rate showed improvement over the preceding year, but it was not 


cited n. 71 
cited n. 71 
cited n. 71 

Sec. II, pp. 37-38. 
Sec. II, pp. 34-35. 
Sec. I, p. 15; quotation, p. 45. 



great. Preventive medicine did not come into its own in this theater until the 
winter of 1944-45. 80 

Logistical Problems in China 

The logistical problems of operating in China and India had a family re- 
semblance. Neither India nor China had a well- advanced industrial structure. 
In both, the American forces were ordered by the War Department to make 
the maximum possible use of local resources of all sorts. In both, the popula- 
tion weighed heavily on the land. In both countries the U.S. forces depended 
on one major line of communications whose operation was perforce left in the 
hands of local authority, since the War Department directed that local re- 
sources be used and American operation would necessarily involve a heavy 
commitment of American resources of men and machines. Despite the family 
resemblance, the logistical problems in China represented the ultimate devel- 
opment of their species. 

India was isolated by distance, but China was blockaded. Inflation was a 
grave concern of the Government of India, but it was out of control in China. 
Indian industry was hard-pressed, but growing and making a major contribu- 
tion to the Allied cause in the Middle and Far East. Chinese industry retro- 
gressed under the pressure of isolation and inflation. The Assam line of com- 
munications had as its principal component a railroad whose capacity had 
never been fully exploited; the eastern line of communications had a 500-mile- 
long road, scratched out of the mountains and navigated by a collection of 
dilapidated trucks. Until 1944, both arteries were the sole responsibility of 
local authority. In India, the American forces drew their food and clothing 
from the Government of India through the SOS. In China, the Americans 
were in the most literal sense guests of the Chinese, for they lived in hostels 
operated by the Chinese War Area Service Corps, known to every American as 

Three logistical problems beyond the control of SOS greatly affected its op- 
erations in China. The Hump may be mentioned first. Suffice it to say the SOS 
problems diminished or increased as the amount of tonnage arriving from 
India fluctuated. The second problem was inflation. Because the U.S. forces 
were directed to live off the country and, in view of China's isolation, had no 
option in the matter, the galloping inflation which steadily weakened the Chi- 
nese war effort made it ever less possible for SOS to obtain goods and serv- 
ices, for example, spare parts and building construction, in the open market. 

The Chinese Government insisted that the exchange rate between Chinese 
and American dollars should be $20 Chinese for $1 American. As of January 
1944, the open or black market rate was 100 to 1. That was bad enough, but 

80 History of India- Burma Theater, 1944-1945, II, 298-300. OCMH. 



prices were rising about six times as fast as the value of the dollar. Notably, 
prices of items used by the Chinese, such as blue cloth and rice, rose far faster 
than those used by Americans. In any event, there were too few Americans in 
China to affect so great a market as the 200 million plus Chinese in unoccupied 
China. The sophisticated and practical Chinese had long before discovered the 
magic of credit, and had an elaborate and nationwide net of small banks or 
"cash shops." Control of these myriad little banks was beyond the administra- 
tive resources of the central government. Speculators borrowed heavily from 
the banks, then accumulated inventories which increased in value far more 
rapidly than even the 9 percent a month the banks were charging by early 1944. 
As a result, from July 1937 to June 1944, the expenditures of the Chinese Na- 
tionalist Government increased 120 times; the purchasing power of its cur- 
rency decreased by 384 times. In terms of actual goods and services, the 
Chinese Nationalist Government effort of 1944 was about one third that of 
1937. Its budget was about 3 percent of the national income. 81 

The eastern line of communications from Kunming to Chennault's east 
China air bases was not under American control. Consistent with the policy 
followed in India, the American authorities had left the matter to the Chinese. 
There were considerable stretches of railway in the eastern line of communica- 
tions that functioned with reasonable efficiency and a high degree of safety. 
The worst bottleneck was the Kutsing-Tushan Highway, a 500-mile stretch, 
and therefore longer than the 483 miles from Ledo, Assam Province, India, to 

Wanting, China. (See Map 1.) 

The majority of the distance is characterized by rugged, mountainous terrain with a prac- 
tically continuous series of hair-pin turns with grades as steep as 27%. The road-bed has a 
crushed rock foundation with a clay soil surface. Much of the road is extremely rough and 
caused a high incidence of spring failures and a rapid deterioration of chassis and bodies. 
The clay surface made travel extremely hazardous during the rainy season. The route has a 
large number of deep cuts and high fills and during the wet season, landslides and cave-ins 
frequently completely blocked all movement for days until native crews could clear the 
obstruction. 82 

In January 1944, the Southwest Highway Transport Administration, the 
principal Chinese Government agency operating on the eastern line of com- 
munications, owned 1,196 two and three-ton trucks of which 183 were oper- 
able. The agency was able to contract for the services, at one time or another 
during the month, of 2,958 privately owned vehicles. Maintenance was of the 
most rudimentary sort. Because inflation steadily cut the salaries of the Chinese 

81 ( 1 ) The Chinese Government finanrprl the hnllf of its war expenditures by issuing paper 
money, as has been noted \v \Stilwell' s Mission to Chinaj (2) SOS in CBI, App. I, SOS in China, Sec. 
II; App. F, Incl 4, Economic Conditions in Free China and Their Effects on Army Procurement. 
(3) Ltr, Col Coke S. Matthews, Pres, Bd of Investigation, to CG, USFCT, 31 Oct 45, with Rpt, 
sub: Housing and Construction for American Personnel and Installations in Kunming, China; 
Exhibit N. CT 40, Dr 4, KCRC. 

82 SOS in CBI, App. I, SOS in China, Sec. II; App. C, p. 4. 



Government employees operating the trucks, they perforce loaded their ve- 
hicles with passengers and goods for private gain. There were no facilities for 
the truck drivers along the route beyond the few little roadside inns. There- 
fore, in February 1944, an average of sixteen trucks a day was moving eastward, 
when ninety should have been the minimum. Wrecked trucks dotted the high- 
way. Not unnaturally, the Chinese trying to operate the eastern line of com- 
munications were thoroughly disheartened. The most hopeful sign was that 
Chennault, his attention now fixed on the eastern line of communications, de- 
manded remedial action. 83 

Though Chennault had been extremely skeptical of the worth of the Ledo 
Road, he had not hesitated in May 1943 to place his air squadrons at the end 
of the much longer eastern line of communications. Not until October 1943 
did he begin to draw the attention of his superiors to the problem of increasing 
its efficiency. The response of CBI Theater headquarters to his pleas and warn- 
ings may justly be compared to its response to the problems of the Assam line 
of communications. 84 

As 1944 began, the SOS in China had a strength of 452 men, suggesting 
the scale of the logistical effort in China. It was organized into Advance Sec- 
tion No. 3 with 354 men, working mostly in and around Kunming, and Ad- 
vance Section No. 4 with 98 men stationed at Kweilin and Heng-yang. The 
sharp disparity in strength reflected the circumstance that the Fourteenth Air 
Force was largely supported by the Air Service Command (ASC). On 20 January 
1944, the two SOS sections merged to become Advance Section No. 1, Col. 
Lewis P. Jordan, commanding. At Kunming and Yun-nan-i were airfreight re- 
ception and discharge stations to handle Hump tonnage. The Ordnance peo- 
ple had a shop at Kunming for 3d and 4th echelon motor maintenance. The 
shop had facilities for manufacturing a few spare parts. There were no ord- 
nance facilities on the eastern line of communications. The 95 th Station Hos- 
pital at Kunming and a hospital improvised by the SOS at Kweilin met 
medical needs. Until March 1944 Chinese nurses were used. Because of the fact 
that each American in China increased demands on Hump tonnage, it was Stil- 
well's policy to keep U.S. personnel in China to the absolute minimum. It was 
believed that American women would need special types of housing and re- 
quire allocation of Hump tonnage to certain modest luxuries such as cosmetics 
which would be better devoted to the stuff of war. 85 

As for subsistence, the War Area Service Corps' support of U.S. operations 
in China was a major project for the Chinese. 86 In the Kunming area the 

" Bykofsky MS. 

84 (1) See pp. 119-20123] above. (2) Memo, Chennault for Stilwell, 30 Dec 43, sub: Transpor- 
tation Facilities from Yunnan Eastwards. The Chennault- Wedemeyer Letter, Item 6. (3) Bykofsky 

85 SOS in CBI, App. I, SOS in China. 

86 ( 1 ) A fuller treatment of WASC is in History of CBI, Section III, Appendix XII, War Area 
Service Command. 



WASC operated thirteen hostels. The hostels were numbered consecutively, 
with one exception. The sentiments of pilots who might find themselves as- 
signed to Hostel No. 13 were respected. Consequently, the numbers ran 11, 
12, and 14. Under U.S. supervision, the WASC operated an abattoir. The Chi- 
nese purchased cattle, hogs, and chickens, and delivered them to the abattoir. 
There they were inspected, butchered, and dressed under supervision of U.S. 
personnel and delivered to the several mess halls in U.S. trucks. The WASC 
furnished butcher, coolies, ice, and everything else needed to operate the abat- 
toir. In the Cheng-tu area, the XX Bomber Command's base, the hostels were 
so far apart that WASC operated five abattoirs and a large co-operative market. 
All laundering was done by WASC. Fresh vegetables and staple products were 
purchased by WASC and delivered to the several mess halls in accord with 
the ration strengths. 

The arrangement was not entirely satisfactory to the United States. Chinese 
intentions were good, but the WASC did not have the administrative ability to 
meet its commitments. Food was often very bad, buildings were not main- 
tained, and sanitary conditions were often below the acceptable minimum. 
WASC padded the daily returns on U.S. personnel accommodated, presumably 
with a view to postwar settlement. By February 1944, the situation appeared 
"critical" to Stilwell. 87 In February, conferences began to see if the United 
States could not become a paying guest in China, with more control over the 
accommodations. The Quartermaster Section did what it could to supplement 
the WASC ration by supplying such items as butter, jam, coffee, powdered 
milk, and sweets, which were flown in from India. Liaison and radio teams in 
the field were given special rations of canned meats, fruits, and vegetables. 

Transportation was hampered because operation of trucks in China was 
most difficult, due to China's isolation, lack of developed oil resources, and 
lack of an industry to supply spare parts in adequate quality and quantity. Re- 
placement vehicles were bought on the open market. The trucks necessarily 
used a locally compounded fuel that was 25 percent gasoline and 75 percent 
alcohol. The alcohol formed acid compounds which attacked the bearings and 
crankshafts, forcing frequent cleaning and repair. The washboard roads over 
which the trucks operated subjected them to a terrible pounding which in- 
evitably shook loose every joint and fitting. 

Construction was principally a Chinese function. Until March 1944, the 
SOS contracted directly with individual Chinese firms for nonoperational 
construction. For operational construction, the SOS prepared plans and sub- 
mitted requests to the Chinese Government. Such projects were paid for by 
the Chinese as part of their contribution to the United Nations war effort. The 
procedure was open to criticism because the Ministry of Finance was reluctant 
to advance the funds even after other Chinese agencies had concurred in the 

87 History of CBI, p. 215. 



In February 1944 a division of responsibility between the SOS and the 
Air Service Command was arranged that gave the latter almost complete 
local responsibility in east China. The SOS would see to it that supplies came 
to Kweilin after which the ASC would take over. That shift in responsibility 
left construction for the instructors with the Second Thirty Divisions, 
transportation operations, and Kweilin hospital the only SOS activities in east 

Very likely one of the major factors behind the new and clear-cut division of 
responsibility was Chennault's representations. The poor operation of the 
eastern line of communications so hampered his operations that in December 
1943 he sent a sharp letter to Stilwell demanding either that the SOS set up an 
adequate line of communications to east China or let him have control over his 
own line of communications. In January 1944, Chennault flew to New Delhi 
to present his case in person. This resulted in the assignment of Lt. Col. 
Maurice E. Sheahan to be transportation officer of Advance Section No. 1. 
Colonel Sheahan's arrival on 2 February marked the beginning of better times. 
Soon after, Headquarters, SOS took more notice of problems in China. Since 
his arrival, Covell's attention had been fixed on India. Not until 23 February 
do his reports to Somervell mention the eastern line of communications, but 
then he describes it as a real problem. 

Surveying the eastern line of communications, Colonel Sheahan found it in 
a depressing state and quickly proposed a series of remedies, dependent of 
course on Hump tonnage. They were grouped together as TIGAR 26 A, or 
TIG-26A. The mission of TIG-26A was to move supplies for the Fourteenth 
Air Force over the line of communications Kunming-Chanyi-Kweiyang- 
Tushan-Liuchow-Kweilin to be distributed around the east China airfields. The 
target for the period July-December 1944 was set at 8,000 tons a month. From 
the termini of Liuchow and Kweilin, Chinese agencies were to move the sup- 
plies. For means, Sheahan wanted to fly in 700 trucks (500 l^-ton 4x2 
Chevrolets and 200 6x6 General Motors) and spare parts, engines, tires, and 
devices to convert gasoline-fueled engines to charcoal burning for the Chinese. 
He also requested three quartermaster truck companies, a heavy automotive 
maintenance company, and a military police platoon, plus overhead to super- 
vise and control the American convoys Sheahan would operate. Simultaneously, 
SOS contemplated TIGAR 26 B, bringing in 500 5-ton truck-trailer combina- 
tions from the Persian Gulf to China via the Soviet Union. Colonel Sheahan's 
proposals began moving up the chain of command. 88 Until they were approved 

88 (1) App. cite dln. (2) Bykofsky MS. (3) See Ch. I, above. (4) Implementation of 
TIGAR 26 B was delayed by the refusal of the Soviet Union to allow vehicles to cross its terri- 
tory. Permission was not granted until fall 1944, and a convoy, code name Lux, was organized in 
the Persian Gulf Command. Uprisings in Sinkiang Province, China, resulted in last-minute 
changes in Lux plans. The Lux movement was changed to enter China via the Ledo Road. With 
the arrival of the Lux convoy in China in March 1945, transportation problems were considerably 



and the means to implement them arrived in CBI, Chennault's operations 
would be limited to what could trickle forward over the eastern line of 


After twenty months' experience proved that the Assam line of com- 
munications was not efficiently run under civilian control, and after the 
Quadrant decisions resulted in service troops being allotted to CBI, it was 
possible to place the control of the line of communications in the hands of an 
Anglo-American military panel, and to use American personnel in key posi- 
tions at Calcutta and on the Bengal and Assam Railway. The result was a great 
increase in tonnage brought forward at a most opportune time, when the 
Japanese were crossing the Indian border. Within the SOS, its administration 
was simplified, and though the Government of India feared that inflation might 
force it to cut back military production, the natural impact of such a move on 
the supplies India could give the U.S. forces on reciprocal aid account (reverse 
lend-lease) remained in the future. There was increasing dissatisfaction with 
the health of the command, but as yet little improvement in the fight against 
the effects of a contaminated environment. In India, the SOS picture in 1943-44 
was a steadily brightening one. 

In China, the picture remained one of problems and projects that outran 
the logistical support at hand. China's isolation, the unchecked inflation that 
was gradually but certainly wrecking her economy, and the extreme difficulties 
of transportation all combined to hobble the U.S. effort. In 1943, the President 
and the Generalissimo had defined the U.S. effort as being in the air, and had 
set it at a level beyond the ability of the CBI Theater to support. That having 
been done, there followed no co-ordinated effort to provide logistical support. 89 
As the spring of 1944 came to east China, the manifold consequences of the 
President's and the Generalissimo's decision were about to unfold. 



Decisions To Attack 

North and east of Burma, diplomatic and military battles of great 
significance for the future of Asia were fought in 1944. Both the Chinese and 
the Japanese resolved on major operations in China Theater. So vast was the 
arena of conflict that the two sides could make their offensive efforts many 
hundreds of miles apart. Unfortunately for the Allies, as they attacked in one 
area and tried to defend in another, they were divided among themselves on 
national lines, and within each nation on lines of policy. The united, resolute 
wholehearted effort that might aid a speedy victory did not exist. 

The Chinese decision to take the offensive ensued upon a complete change 
in the President's point of view. Though the United States and the British 
Commonwealth at Cairo in December 1943 had gone back on their long- 
standing and oft-repeated commitments for a major amphibious operation in 
the Bay of Bengal, the President now supported Marshall and Stilwell in taking 
the line that China's self-interest required the Chinese to attack across the 
Salween River into Burma. This change in Roosevelt's thinking meant support 
for the thesis long advanced by the War Department, that the Chinese should 
be asked to take definite aggressive action with the lend-lease, credits, and air 
support they received from the United States. With Stilwell fully occupied in 
north Burma, it was the President who took the lead in calling on the General- 
issimo to act, and with messages that had clear echoes of the bargaining or 
quid pro quo approach so long urged by Marshall, Stimson, and Stilwell. 

" Money Is the Root of All Our Trouble" 

The President's radio of 29 December 1943 was strong in tone, explaining 
that considerable equipment and instructor personnel were scheduled for move- 
ment to China to assist in preparing the Y-Force for Burma operations. The 
President told the Generalissimo that he wanted to "avoid at this time the use 
of the restricted airlift in the employment of these resources in an effort, the 
full impact of which upon the enemy will be delayed." 1 

With Stilwell in Burma, his own Chief of Staff, China, Burma and India 
Theater, General Hearn, had the duty of delivering the President's messages to 

1 (1) Memo 103, Roosevelt for Gpnpra ljs.sim o. 29 Dec 43 (delivered by Hearn). Item 1578, 
Bk 5JWS Personal File. (2) See Chsjl^ancpv] above. 



the Chinese. This, however, did not mean that General Hearn was in any sense 
a staff officer of the Generalissimo's China Theater. Hearn informed Stilwell 
of the contents of each of Roosevelt's messages and relayed StilwelPs occa- 
sional comments on them to whichever Chinese dignitary, usually Madame 
Chiang, might accept them for the Generalissimo. 

Noting the President's changed attitude, Stilwell told Hearn to impress on 
the Chinese that if Y-Force did not co-operate in Burma operations, the Presi- 
dent might lose interest, and might conclude that the Chinese themselves were 
not really concerned about breaking the blockade of China. Stilwell also sug- 
gested the Chinese be told to weigh the effect on themselves of SEAC's 
suggestion that Burma be bypassed. 2 

The President himself, on 14 January 1944, finally began to hint that lend- 
lease to China might cease if it was not to be used against the Japanese. 
Beginning with the remark that the opening of the Burma Road was " . . the 
next and most immediate solution to our present problem," Roosevelt went on 
to tell the Generalissimo: 

1 am informed that the Ledo Road forces are trained, equipped and in position against 
the enemy in Notch Burma, and that the progress of the Ledo Road secured by these troops 
is making good headway. I am of the opinion therefore that all of us should concentrate our 
efforts with the means at hand to push vigorously all military operations as will assist this 
road project. Mountbatten's plan and extent of operations depend in large measure, as you 
are aware, upon support from Yunnan. I know that you are in agreement therefore that it is 
most important that all possible pressure with available means be exerted by your Yunnan 
forces in coordination with Admiral Mountbatten's operations from India. If the Yunnan 
forces cannot be employed it would appear that we should avoid for the present the move- 
ment of critical materials to them over the limited lines of communication and curtail the 
continuing build-up of stockpiles in India beyond that which will be brought to bear against 
the enemy. 3 

The President's messages and Hearn's comments on them made their 
impression on Madame Chiang. On 7 January she remarked to Hearn that the 
Generalissimo had not answered the President's December messages because 
Roosevelt had not yet agreed to the Generalissimo's request that the United 
States lend China one billion dollars in U.S. currency or pay for the Cheng-tu 
airfields at the 20 to 1 rate. General Hearn at once concluded that the Chinese 
were linking financial aid with operations in Burma. 4 

General Hearn was not alone in thinking so. In faraway Washington, the 
memorandums that passed among the Treasury, War, and State Departments as 

2 (1) Ltr, Hearn to authors, 16 Feb 50. OCMH. (2) Rad RELOT OT 80, Hearn to Stilwell, 21 
Dec 43; Memo 104, Hearn for Mme. Chiang, 29 Dec 43. Items 1552, 1585, Bk 5, JWS Personal 
File. (3) Memo, Stilwell for Hearn, 11 Jan 44. Item 263, Bk 3, JWS Personal File. Hearn permitted 
Madame Chiang to read this memorandum. 

5 CM-OUT 4277, Roosevelt to Generalissimo, 14 Jan 44. Item 1629, Bk 5, JWS Personal File. 
The text of this message was presented to the Generalissimo as Memorandum 118 on 15 January 
1944. _ 

4 (1) See |Ch. Il| above. (2) Rad RELOT 9, Hearn to Stilwell, 9 Jan 44. Item 1608, Bk 5, JWS 
Personal FileT^™^^ 



they followed the negotiations of their Chungking representatives revealed 
similar reactions, and disclosed a growing feeling among their authors that the 
Chinese, by placing a dollar value on their share in the war, were overreaching 
themselves. Many of these officials believed the President shared their views. 5 

The Generalissimo's response to Roosevelt's December messages (the 
President's latest having crossed it in transmission) confirmed the impressions 
mentioned above. In tone, it was a virtual ultimatum. The Generalissimo stated 
that unless the billion dollars was forthcoming or unless the United States 
agreed to finance the Cheng-tu project at the rate of 20 to 1, the Chinese would 
no longer provide food and housing to the U.S. forces in China after 1 March 
1944. The Generalissimo told the President that he had "even gone to the 
length of delaying the reopening of the Burma route so that essential amphibi- 
ous equipment should be diverted to the European theater, thereby disappointing 
all classes of my countrymen." 6 

Sent to Washington through State Department channels, the Generalissimo's 
message met a mixed reception in Washington. The first draft of the United 
States answer, as prepared by Somervell, assured the Generalissimo that con- 
trary to his charge the exchange transactions proposed by the United States 
were not commercial in nature; that while the United States was prepared to 
pay all expenses incurred by the U.S. Army it refused to do so at a fixed rate of 
20 to 1 in a money market that was racing upward; that any such insistence by 
the Chinese would not be understood by the American people nor by the 
Congress; and that the President must insist on some reasonable arrangement 
for Chinese currency, either by outright donation or some form of reverse 

The draft was shown to Marshall and Arnold on the ground that its dispatch 
might well lead to a break in Sino- American relations and the withdrawal of 
the United States forces from China. Marshall and Arnold approved the paper. 
It then went to the President who also approved, subject to the later concur- 
rence of the Secretary of State, Cordell Hull. 7 Hull demurred, urged the Presi- 
dent to reconsider, and his more cautious approach carried the day. The 
President and his advisers drew back from risking an open breach with China, 
but that the highest political and military authorities in the United States were 

5 (1) Teletype Transcript, Conv between Morgenthau and Somervell, 15 Jan 44. Memo with 
incls, Somervell for McNarney, 1 5 Jan 44; Rad, Morgenthau to Adler, Treasury Dept Representa- 
tive in Chungking, 15 Jan 44. Somervell File, Vol IV, CBI 1944. (2) Memo, Somervell for 
Marshall, 15 Feb 44, sub: Exchange Policy With China. Somervell China File, A47-81. In this 
memorandum, Somervell reviews problems of the exchange rate. 

6 (1) Rad STATE 105, Generalissimo to Roosevelt, 16 Jan 44 ABC 336 (China) 26 Jan 42, 
Sec 1A, A48-224. (2) Rad RELOT 26, Hearn to Stilwell, 18 Jan 44. Item 1640, Bk 5, JWS 
Personal File. 

7 (1) Draft, Suggested Msg from President to Generalissimo, undated. Somervell China File, 
A47-81, and Somervell File, Vol IV, CBI 1944. (2) CM-OUT 8054, Somervell for Marshall 
to Stilwell, 20 Jan 44. (3) CM-OUT 8256, Somervell to Stilwell, 21 Jan 44. (4) Memo, Lt 
Col Paul G. Pennoyer, Asiatic Sec OPD, for Col Charles H. Caldwell, Asiatic Sec OPD, 20 
Jan 44, sub: Chinese Currency Negotiations. ABC 336 (China) 26 Jan 42, Sec 1A, A48-224. 
Colonel Pennoyer followed the financial negotiations for OPD. 



willing to contemplate such action dramatically illustrated the change in 
China's diplomatic position vis-a-vis the United States. 

An anxious ten-day interval followed while the United States and China 
considered their next moves. The silence was broken by Ambassador Gauss's 
delivery of the U.S. reply to the Chinese ultimatum. In it, the President 
remarked that he could not . . escape the feeling that because of the distance 
between us there may be danger that we may fail adequately to work out our 
common problems and may rush into decisions which would not be in the 
interests of either of our peoples." Roosevelt suggested that the Chinese send 
a representative to Washington to discuss the exchange rate with Treasury 
officials. There was no mention of the billion-dollar loan. Then came an indi- 
cation of the President's attitude towards China. Roosevelt flatly stated that 
beginning 1 March 1944 U.S. Army expenditures in China would be limited 
to U.S. $25,000,000 a month. 8 

As of 15 March 1944, the open-market rate for the U.S. dollar in Kunming 
(rates varied from city to city) was 230 to 1. If the Chinese Government insisted 
on making its dollars available at only 20 to 1, U.S. expenditures in March, for 
example, would be equal to what would be permitted by purchasing U.S. 
$2,000,000 worth of currency in the open market. Since the actual exchange rate 
was working ever upward, Chinese insistence on 20 to 1 would ultimately force 
American units limited to U.S. $25,000,000 a month to withdraw from China. 
Conversely, if the United States acquiesced in the Chinese demand for the 
20 to 1 rate, while the Chinese continued to finance their war by paper money, 
and the Chinese banks continued to lend to speculators who drove up the price 
level, the United States would ultimately be spending its dollars by the billion 
merely to clothe and feed a few thousand Americans in China. 9 

When the Generalissimo replied to the President, his message of 2 February 
linked China's financial and military problems: 

I have received your message dated twenty-sixth January transmitted by Ambassador 
Gauss and I am deeply appreciative of your efforts to help me and my government. I have 
consulted with Dr. Kung [Minister of Finance] regarding the suggestions contained therein 
and have requested him to acquaint the Ambassador and General Stilwell's representatives 
with the decisions he and I have agreed upon [the Chinese were now offering 30 to 1]. I 

8 Messages exchanged during this ten-day period among State, Treasury, and War Depart- 
ment representatives in Chungking and officials in Washington, including the President, are 
in Somervell's China File, A47-81. The contents of Somervell's file suggest that it was his job 
to co-ordinate all policy statements among the interested departments after they had received 
and circulated the Generalissimo's ultimatum of 16 January 1944. Though the President's reply 
was sent to Chungking on 20 January 1944 as CM-OUT 8054, it was subsequently altered and 
was given to the Generalissimo on 26 January by Ambassador Gauss. Two days later Gauss, 
Hearn, and Treasury representatives were received by H. H. Kung, who had been designated 
by the Generalissimo to discuss the details of President Roosevelt's proposals. The 20 January 
text of the President's message and subsequent changes are in CM-OUT 8054, Roosevelt to 
Generalissimo, 20 January 1944, and CM-OUT 9567, Somervell to Hearn, 24 January 1944. 

9 Memo, Dr. Edward C. Acheson, Financial Adviser to CG, USAF CBI, for Director or Material, 
Hq ASF, 20 Mar 44, sub: Course of Financial Negotiations in China, Rpts 3 & 4 (Feb-Mar 44). 
Somervell China File, A47-81. 



trust that very shortly a solution satisfactory to both our countries may be reached. I wish 
to assure you that Dr. Kung and I have exerted our utmost to meet your wishes short of 
jeopardizing China's economic front to the breaking point, and short of endangering the 
morale of our people in the prosecution of continued resistance. 

I wish also to acknowledge your telegram sent through General Stilwell's headquarters 
dated fifteenth January. I appreciate your desire to open up the Ledo Road, a desire which is 
also my great concern since it is only through the opening of this land route that China may 
quickly obtain the heavy equipment much needed by her army. You doubtless recall that at 
Cairo I reiterated and emphasized the fact that I am ready to send the Yunnan troops to 
Burma at any moment that large-scale amphibious landing operations can be effected at 
strategic points. I stand ready to adhere to this decision, and hope that we can carry out 
operations even before November of this year, which date you mentioned as possible and 
probable for the diverting of the amphibious equipment to Burma. 

I am leaving for the Hunan front tomorrow and shall be away for a fortnight. Any 
message will forwarded. I know you realize that the year will prove a most critical period for 
China both in the economic and military sense, but I am confident that with your help we 
shall pull through. . . . 10 

The President's acknowledgement of this message was limited to assurances 
that the United States would continue to study both exchange rates and U.S. 
military expenditures in China. 11 

The foregoing communications from Washington had their effect within 
the Chinese Government. Apparently on his own authority as Minister of 
Finance, Dr. H. H. Kung undertook to advance Chinese National currency 
(CN) $15,000,000,000 to CBI Theater headquarters to cover U.S. expenditures 
for March, April, and May 1944. The credit was extended with no stipulation 
as to the rate of exchange for repayment. Of this credit, Hearn received an 
immediate grant of CN $500,000,000 for the Cheng-tu fields. There was, how- 
ever, one obstacle to the prompt use of these funds. The actual currency 
involved was Chinese paper money, engraved in the United States and stock- 
piled in India. It would have to be flown over the Hump, like everything else 
the U.S. forces in China used. 

The need of allocating Hump tonnage for these currency advances brought 
now familiar repercussions in China. Where air cargo was concerned Chinese 
currency was measured as tonnage rather than by dollar value. The bills in 
India were of small denominations lest they add to the general inflationary 
psychology. Experience revealed that a ton of currency might vary in value from 
CN $7,000,000 to CN $100,000,000. The Fourteenth Air Force complained that 
4,500 gallons of gasoline were used in one day's lift of Chinese currency. Hearn 
tried to placate the Fourteenth by explaining that the chore was a necessity, 

10 Rad AGWAR 79, Generalissimo to Roosevelt, 2 Feb 44. Item 1697, Bk 5, JWS Personal 
File. Hearn and Gauss doubted that the Generalissimo was touring the "Hunan front" since they 
believed none existed. Hearn reported that the Generalissimo and his wife desired to vacate 
Chungking for Hunan for a short rest and to avoid decisions on the exchange rate. CM-IN 
3024, Hearn to Somervell, 5 Feb 44. 

II Rad WAR 4457, Roosevelt to Generalissimo, 7 Feb 44. Item 1714, Bk 5, JWS Personal 



that each haul had to be charged against operational tonnage. Casting his eye 
on some of the less apparent aspects of war in China, he remarked to 
Chennault: "Money is the root of all our trouble." 12 

American Military Observers in North China 

On 9 February 1944, the President raised a point with the Generalissimo 
which was part of an issue whose eventual outcome was possibly beyond the 
power of any man then living to foresee. And it was to grow until it dominated 
the diplomacy of the Pacific Ocean as once the problem of Japanese imperialism 
had done. This was the Chinese Communist issue. The episode began quietly 
enough, with a suggestion by Stilwell's political adviser, John P. Davies, Jr. 

In 1943, on 24 June, Davies had suggested to the State Department that a 
consulate general be opened in the territory of the Chinese Communists, and 
that a military mission be sent to them. 13 Stilwell's papers do not mention the 
project; his attitude then is unknown. 

At the Sextant Conference, Davies had been present when Stilwell 
attempted to get some policy guidance from the President. It may be that 
Davies remembered the President's reaction to Stilwell's warning that a repeti- 
tion of the Japanese Tung- Ting Lake operations of May 1943 might overthrow 
the Generalissimo: "Well, then we should look for some other man or group 
of men, to carry on." 14 In any event, twenty-five days later Davies began to 
send Hopkins a series of memorandums on Chinese domestic matters, which 
Hopkins in turn relayed to the President. 

In his first memorandum, of 31 December 1943, Davies began by observing, 
"The Generalissimo is probably the only Chinese who shares the popular 
American misconception that Chiang Kai-shek is China." He closed with the 
recommendation: "A realistic policy towards Chiang would be based on (1) 
recognition by us that the Generalissimo is highly susceptible to firm 
co-ordinated American pressure, (2) stern bargaining (in consultation with 
American representatives in China), and (3) readiness to support a strong new 
coalition offering cooperation mutually more beneficial to China and the 
United States." 15 

Davies a fortnight later returned to the suggestion he had made in June 
12 History of CBI, Sec. Ill, Ch. XI, China Exchange. 

"Memo by Davies, Second Secy of Embassy detailed to Gen Stilwell, 24 Jun 43, sub: The 
American Stake in Chinese Unity: Proposals for Preliminary American Action. History of CBI, 
Sec. Ill, Ch. I, Item VI. 

14 (1) See Ch. II, above. (2) The Stilwell Papers, on page 252, quotes Roosevelt to that 

15 Memo, Davies, New Delhi, 31 Dec 43, sub: Chiang Kai-shek and China. On 7 February 
1944 Hopkins sent the memorandum to the President with the note: "Here is another interest- 
ing letter I got from John Davies Jr., who was with our Embassy in China and is now with 
Stilwell and Mountbatten." Bk IX, Hopkins Papers. Hopkins' note would suggest that Davies 
had written even earlier. However, such correspondence was not seen by the authors, and Davies' 
own covering letter suggests this is the first of a projected series. See Ltr, Davies to Hopkins, 
31 Dec 43. Bk IX, Hopkins Papers. 



1943, of a U.S. military mission to the Chinese Communists. After drafting his 
suggestions on 15 January 1944, he made a visit to Stilwell's headquarters in 
north Burma on the 20th. What Stilwell thought of the project Davies did not 
record, but presumably he approved. On 23 January Davies suggested to 
Hopkins that the President arrange for Stilwell's headquarters to send U.S. 
observers to Communist China. Davies offered Hopkins several reasons for the 
presence of such a mission in and around the Communist base at Yenan: 

We need to despatch immediately while it is still welcome, a military and political 
observers' mission to Communist China to collect enemy information, assist in and prepare 
for certain limited operations from that area, obtain accurate estimates of the strength of 
Communist Armies, report on Russian operations in North China and Manchuria should 
Russia attack Japan, and assess the possibility of North China and Manchuria developing 
into a separate Chinese state— perhaps even as a Russian satellite. Chiang's blockade of the 
Communists and their consequent isolation, are forcing them toward dependence upon 
Russia. An American observers' mission would break this isolation, reduce the tendency 
toward dependence upon Russia and, at the same time, serve to check Chiang's desire to 
attempt liquidation of the Communists by civil war. The Generalissimo will naturally be 
opposed to the dispatch of American observers to Communist China. His permission cannot 
be obtained through ordinary diplomatic and military channels. The request should come to 
him directly from the President, who can overcome any initial refusal by exercise of our 
ample bargaining power. 16 

A contemporary example of such a mission to a Communist faction within 
an Allied state existed, for when Davies wrote his letter a British military mis- 
sion under Brigadier Fitzroy Maclean had been for several months with Marshal 
Josip Broz Tito in Yugoslavia, and Maclean had been at Sextant when Davies 
was there. 17 

Apparently the President liked the suggestion, for on 9 February he asked the 
Generalissimo if an American observer mission might go to north China. 18 The 
request arrived at an inauspicious moment. It nearly crossed a report from 
Stilwell to Marshall that a crisis in Nationalist-Communist relations might be 
at hand. Stilwell understood that the Nationalist armies facing the Communists 
in north China had been built up to 500,000 men, that the Nationalists were 
accumulating supplies for this force, while the Communists for their part were 
reinforcing. Stilwell thought it unlikely that there would be a civil clash before 
the end of the war with Japan. Reminding Marshall that this mutual hostility 
in north China was hindering the war against Japan, he reported that for 
obvious reasons the Japanese Domei News Agency was goading both sides. 19 

As predicted by Davies on 23 January, the Generalissimo refused the 

16 Ltr, Davies to Hopkins, 23 Jan 44, with Incl, Memo, 15 Jan 44, sub: Observers' Mission 
to North China. Bk IX, Hopkins Papers. 

17 Fitzroy Maclean, Escape to Adventure (Boston. Little, Brown & Company, 1950). 

18 (1) A copy of the original 9 February radio is in Book IX, Hopkins Papers. (2) In 
Chungking, the President's radio was copied as a memorandum from General Hearn to Madame 
Kung on 19 February 1944, and presented to her with the request that she give the President's 
radio to Madame Chiang for the Generalissimo. Item 1726, Bk 5, JWS Personal File. 

19 CM-IN 2373, Stilwell to Marshall, 4 Feb 44. (Probably originated by Hearn.) 



President's request, in polite terms, by agreeing that such an American military 
mission might visit any territory in north China which acknowledged the 
authority of the Nationalist Government. The President accepted gracefully 
and stated that such a mission would be dispatched in the near future to north 
China. 20 The President's reply did not mention the Generalissimo's limits on 
such a mission's movements, and very tactfully he soon began to persuade the 
Generalissimo to remove them. 21 

An indication of the attitude the Generalissimo's government might 
contemplate adopting toward the Chinese Communists was offered when his 
Chief of Ordnance, Gen. Yu Ta-wei, asked Hearn if lend-lease arms could be 
used against the Communists. Hearn was understood by Ambassador Gauss to 
have replied that the Chinese could not successfully explain such an action to 
the people of the United States. 22 

SEA C and Stilwell Obtain Pressure on Chiang 

Though after January 1944 there were several major points at issue between 
them regarding Allied strategy for Southeast Asia, Mountbatten and Stilwell 
agreed that Y-Force should hasten to join the Burma campaign, and both 
wanted the President and the Prime Minister to put pressure on the Generalis- 
simo to that end. Mountbatten and Stilwell had discussed their personal rela- 
tionship on 4 March and patched up their differences. Then the long-expected 
Japanese attack on India had begun. The diversion of Hump transports it had 
caused and the obvious menace which it posed placed new strains on Anglo- 
American strategy in Southeast Asia. 

Stilwell responded to the situation by suggesting to Mountbatten that an 
advance by Y-Force would create a major diversion in the Japanese rear area 
which would be the answer to SEAC's problems, and, by implication, to 
Stil well's immediate troubles as well. Could not Mountbatten place the matter 
before his superiors and ask them to use their good offices with the Chinese? 23 
A day later Stilwell sent a similar request to the Operations Division of the 
War Department, with an information copy to General Marshall. Vividly and 
strongly, he told the War Department that if ever he needed help, now, right 
now, was the time. Pressure on the Generalissimo to move, pressure from the 
highest quarters, was desired. If the Y-Force took the old jade center of Teng- 

20 Memo, Generalissimo to Hearn for President, 22 Feb 44; Memo, Hearn to Generalissimo 
for President, 2 Mar 44. Items 2023, 2054, Bk 6, JWS Personal File. 

21 Memo, Hearn to Generalissimo for President, 22 Mar 44. Item 2126, Bk 6, JWS Personal 

22 Copy, Memo for Admiral Leahy, no date, sub: U.S. Observers Mission to North China. 
Initialed "VE" (Vernon Evans), this copy was in the G-3 files of Hq CBI Theater, and was 
brought back to OCMH with other miscellaneous documents by Col. Harry L. Mayfield, former 
India-Burma Theater Historian, in August 1946. KCRC. 

2} (1) See Ch. IV, above. (2) Rad FG 63, Stilwell to AMMDEL, 16 Mar 44. Item 71, Bk 
6A, JWS Personal File. 



chung, well north of the Burma Road, the Ledo Road could take that route and 
bypass the heavy Japanese concentration on the Burma Road. 24 

Stilwell's suggestion met with Mountbatten's hearty approval. General 
Sultan, Stilwell's deputy in New Delhi, was asked to relay SEAC's approval to 
Stilwell, while down in Ceylon the SE AC staff began to prepare an appropriate 
message to London and Washington. 25 As sent, the SEAC radio placed 
Mountbatten's command squarely behind Stilwell's views on Y-Force: 

I have discussed with General Stilwell question of advance by Chinese forces in Yunnan. 
We agree immediate advance would be of value because it will contain 36 Division and may 
cause diversion of other troops which would otherwise be used against Wingate or Stilwell 
or support offensive against 4 Corps front. We both feel that only a personal approach from 
the Prime Minister and the President through their representatives in Chungking is likely 
to succeed with Generalissimo. Line might be taken that Ledo Force is already achieving 
great success, while presence in Burma of Wingate's forces opens up possibility of victory 
in Imphal area where enemy may find himself committed to an offensive which he cannot 
sustain with his rear threatened. Participation of Chinese from Yunnan in victory which it is 
hoped to achieve would be of great value to morale of Chinese Army and people. Their 
chance of success is considerable since they will initially only have to overcome one division, 
at present holding a very extended front, and reinforcements can only be provided by Japa- 
nese at expense of forces already committed on other fronts. On the other hand position of 
China in eyes of the world is bound to suffer if only Chinese force which has taken part in 
victory is that trained and led by Americans. I should be most grateful if an approach could 
be made with extreme urgency. 26 

Mountbatten's urgings brought a quick response, for the President 
addressed the Generalissimo two days later, and used all of Mountbatten's 
arguments. In sequence, Roosevelt reviewed the "magnificent" pounding the 
Chinese Army in India was giving the Japanese 18th Division, the implications 
of the Arakan and Imphal fighting in tying down the major portion of the 
Japanese strength, and the Chindits' threat to the Japanese supply lines. The 
President warned that if the Allies failed to seize the opportunity thus pre- 
sented, the Japanese might recover and take the offensive. Showing his grasp 
of the North Burma Campaign, the President predicted that the 18th Division 
would require reinforcements and that they would be obtained by withdrawing 
a regiment from the %th Division along the Salween. The President wanted the 
Generalissimo to have these views "at length and in considerable detail in the 
hope that you will give orders to the commander of your Yunnan force to 
cooperate in developing what appears to be a great opportunity." 27 

After presenting the President's message to the Chinese, Hearn urged 
Stilwell to fly to Chungking because there appeared to be signs the Chinese 

24 CM-IN 11455, Stilwell to Marshall, 17 Mar 44. 

25 Rad RE 473, Sultan to Stilwell, 17 Mar 44. Item 75, Bk 6A, JWS Personal File. 
26 Rad SEACOS 116, Mountbatten to COS and JSM, 17 Mar 44. SEAC War Diary. 
27 CM-OUT 4762, Roosevelt to Generalissimo, 17 Mar 44. Hearn relayed this message as 
Memorandum 163 to the Chinese. Item 2109, Bk 6, JWS Personal File. 



were willing to send two divisions to reinforce the Chinese Army in India. 28 
When he conferred with the Generalissimo on 28 March, Stilwell found the 
Chinese had answered the President. 

The Generalissimo Warns of Trouble 

By coincidence, the Generalissimo had chosen to address the President on 
17 March, and naturally his radio crossed the President's response to Mount- 
batten's pleadings. The Chinese leader told Roosevelt: 

For your personal information, I should like to advise you of recent significant develop- 
ments which are matters of grave concern to the prosecution of the war in the Far East. 

First, on the 11th instant while Chinese troops stationed at Sinkiang were engaged in 
suppressing bandits at Hopan, a place situated between Chengkwa and Kitai about 70 
kilometers from the borders of Outer Mongolia, they were bombed and machine-gunned 
twice by planes which flew over from the direction of Outer Mongolia. The first batch con- 
sisted of two planes and the second ten. They all bore the soviet red Star insignia. On the 
12th, planes bearing the same insignia came twice and dropped bombs. On the 13th, they 
re-appeared and machine-gunned. This cannot be construed as a local incident, but is a very 
significant indication of the Soviet Far Eastern policy both now and in the future. 

Second, though the Chinese Communist party have outwardly professed support of the 
Chinese Government's policy of resistance against Japanese aggression, since February they 
have been secretly assembling their guerrilla units from various places and concentrating 
them in North Shensi, evidently preparing for an opportune moment to rise in revolt and 
take Sian [Hsian], the base of our operations in the Yellow River valley. The indications are 
manifest. Considering the matter objectively, it does not seem likely that the Chinese Com- 
munist party would dare to make such a move without some understanding having been 
reached between the Soviet and the Japanese. 

Third, I have information that the Japanese in the near future will launch a large-scale 
offensive in the Chengchow-Loyang area on the Hankow-Peiping railroad line. The enemy 
is moving troops from Manchuria for this purpose. 

Fourth, regarding definite intelligence reports of the Japanese navy, I shall shortly 
forward same to you. 29 

Having dispatched this warning, the Generalissimo then found himself 
obliged to answer the President's 17 March radio. This he did on the 27th. He 
said bluntly that American observers could not visit Communist territory, and 
said it was "impossible" to have an offensive from Yunnan. In his previous 
discussion of affairs in China, the Generalissimo had referred to a Japanese 
ground offensive in muted tones. Now, the topic disappeared into the general 
assertion that China should not launch an offensive into Burma because she 
was too weak: 

The situation in China theater has recently become so grave that I deem it imperative to 
acquaint you with it. The state of affairs in Sinkiang has become very tense since its invasion 
by Soviet planes and Outer Mongolian troops about the middle of this month, with the 
result that our army and people's belief in concerted action by the United Nations has been 

28 Rad CFB 15004, Hearn to Stilwell, 19 Mar 44. Item 2117, Bk 6, JWS Personal File. 
29 Rad AGWAR 184, Generalissimo to Roosevelt, 17 Mar 44. Item 2110, Bk 6, JWS Personal 



somewhat shaken; in other words, our army and people are beginning to ask themselves 
whether the United Nations' pledges and declarations still hold good. 

At the present moment what China can possibly do to fulfill her obligation to the Allies 
as well as to discharge her duty to herself are: (1) to devote all her energy and resources to 
the maintaining of the various fronts in the China theater against any surprise attack by the 
enemy, this theater being the only land base from which to bomb Japan Proper on a large 
scale and with certain effect; and (2) To prepare herself for the day— may it not be distant— 
when Allied land and naval forces can be dispatched to the China coast and the Chinese 
Army can co-operate with them in consolidating our position in East Asia, which will be an 
important base for the invasion of Japan. These are the most important tasks which have 
devolved on China today and which constitute at the same time an obligation she has 
assumed to the Allies. 

In this connection it may be observed that seven years of war have taxed China's material 
and military strength to such an extent that to insist upon her doing something beyond her 
power would be to court disaster, the consequences of which would seriously affect not only 
Yunnan and Szechwan, but also the whole situation in this theater of war. Should this 
happen, the Japanese would invade Yunnan and Szechwan, the revolt in Sinkiang and the 
communists activities in Shensi would assume a new aspect in furtherance of their plan of 
bolschevizing [sic~\ this country so that our Government would not be in a position to do 
its part in this global war, and the Allies in East Asia would be deprived of a base of 
operations against Japan. 

For these reasons, and bearing in mind our obligation to the Allies and our duty to our- 
selves, I am of the opinion that as long as our line of defences has not been adequately 
strengthened, it is impossible for our main forces to undertake an offence from Yunnan. In 
the course of our conversations at Cairo I told you that as soon as the British began large 
scale amphibious operations along the Burma coast, our main forces would launch a vigorous 
attack on Burma with all their might. That promise will be made good when the time 
comes. I realize that reinforcements should be sent to Burma in view of the military situation 
there and that although this does not fall within scope of our work, still we should do what 
we can in compliance with your request. I have therefore decided to dispatch to India by air 
as many of our troops in Yunnan as can be spared in order to re-enforce the troops in Ledo, 
thus enabling the latter to carry on their tasks of defeating the enemy. 

In conclusion I may add that so far as land operations in East Asia are concerned, China 
bears a very heavy responsibility; and, appreciative of the kind and sympathetic assistance 
you have been rendering her all these years, she is determined to discharge that responsibility 
to the best of her ability. I have explained to you quite frankly the present situation in this 
theater of war and the plan of coping with it, in the hope that you will continue to place 
confidence in my country and in one who is your good friend. China, on her part, will not 
fail to do her utmost in the discharge of her responsibility vis-a-vis the Allies. 30 

It must be noted that the Generalissimo's objections were of the most 
general nature. Possibly he felt that his 29 April 1943 pledge to the President— 
"The Generalissimo also wishes me [T. V. Soong] to transmit to you his per- 
sonal assurance that in the event the enemy attempts to interrupt the air 
offensive by a ground advance on the air bases, the advance can be halted by 
the existing Chinese forces"— would cause an intolerable loss of face if for any 
reason he hinted that he could not defend the east China airfields. 31 Or again 

30 (1) CM-IN 21368, Stilwell to Marshall, 30 Mar 44. (2) Msg, Generalissimo to Roosevelt, 
27 Mar 44 (sent as Rad CFB 15407. 29 Mar 44). Item 2145, Bk 6, JWS Personal File. 

31 Stilwell 's Mission to C^/w^ JTnapterT^l quotes the full text of the Generalissimo's 23 April 
1943 pledge to Roosevelt. 



it may be that he saw no specific reasons for alarm. Certainly a SEAC 
amphibious assault on the Arakan coast could not remedy a crisis in east China. 

Whatever his reasoning, the Generalissimo's radio was not well received 
in Washington. The President had been following the course of the North 
Burma Campaign with close attention in the White House map room. He saw 
the pins that marked the Chinese Army in India move deeper into Burma. On 
the far side of the map he saw the massed array of pins that represented the 
Y-Force along the Salween. Facing the Y-Force were the few scattered red pins 
that marked the Japanese 36th Division. The President saw a great strategic 
opportunity, and simply could not understand why the Generalissimo was so 
hesitant to attack one Japanese division. 32 

When in December 1941 Churchill had come to Washington, he found the 
President and his closest advisers believing that China was a great military 
power and classing her armies with the Russian. 33 Now, in 1944, the Gener- 
alissimo was reluctant to send a whole group of these armies, which the Presi- 
dent had rated on a par with the Russian, against the 36th Division. 

What the map room pins did not show, but what the Generalissimo may 
well have had in mind, was the current state of the Y-Force. In the last year, 
considerable progress had been made in supplying artillery and ammunition to 
that force, and by 20 April the Chinese had been given, for example, 244 75- 
mm. pack howitzers. But the Chinese had never brought the Y-Force divisions 
up to strength, either by supplying replacements or by merging understrength 
units. With regard to the intangible but vital elements of training and morale, 
the picture was uncertain. The Chinese staffs and services of supply did not in- 
spire confidence in those Americans who worked with them, for they did not 
appear up to their professional responsibilities. The underlying reality of the 
situation was that the Generalissimo was being urged to attack about 11,000 
well-trained and well-led Japanese and 36 guns with 72,000 Chinese of indif- 
ferent quality whose several headquarters had been issued 244 American can- 
non in addition to their Chinese ordnance. 34 

The President Demands Action 

The first reaction from the President to the Generalissimo's excusing China 
from any offensive effort was a mere formal acknowledgment. Then in the next 
few days a message from Stilwell to Marshall arrived in Washington describ- 
ing the situation created by the Japanese attack on India in somber tones to 
prepare higher authority for a possible success of the Japanese offensive. 35 In 

"Marshall Intervs, 6, 13 Jul 49. OCMH. 

33 Winston S. Churchill, The Hinge of Fate (Boston. Houghton Mifflin Co., 1950), p. 13. 

34 (1) A more detailed description of the Y-Force is given on pages 333-34, below. (2) SNF 
215 has several rather disparaging reports on the Y-Force, dated October 1943. (3) The 56th 
Division had seven and a half battalions in May 1944. (Japanese Officers' Comments.) Its strength 
is given in Japanese Study 93, p. 41. 

55 ( 1) Rad CFB 15442, Stilwell to Marshall, 30 Mar 44. Item 2146, Bk 6, JWS Personal File. 
(2) Rad, Stilwell to Marshall, 31 Mar 44; Rad TK 24, Stilwell to Marshall, 4 Apr 44. Items 
98, 103, Bk 6A, JWS Personal File. 



all probability, the grave prospects for the Allied cause in Asia which this mes- 
sage placed before the Chief of Staff and the President led to the drafting of a 
note from the President in language that showed Mr. Roosevelt's attitude had 
completely changed from his complaisance of March 1943. His radio of 3 April 
1944 said: 

The present offensive by the Japs in the Imphal area is directed primarily against the line 
of communication which makes possible the transportation of materials to China. If the 
Japanese succeed in their intentions in this drive, they can then concentrate against and de- 
stroy the Ledo Force and turn against your Yoke Force at their leisure. 

The British are ably meeting the strong Japanese threat to the line of communication to 
China and the supply route which supports your troops in the Mogaung Valley. 

While heavy fighting is in progress in West Burma and on the Arakan Coast, the Sal- 
ween front has remained quiet and as a result the Japanese have been able to divert ele- 
ments of the Fifty-sixth Division to meet StilwelPs thrust down the Mogaung Valley and the 
threat of the Long Range Penetration Groups in North Burma. It is inconceivable to me 
that your Yoke Forces, with their American equipment, would be unable to advance against 
the Japanese Fifty-sixth Division in its present depleted strength. To me the time is ripe for 
elements of your Seventy-first Army group to advance without further delay and seize the 
Tengchung-Lungling areas. A shell of a division opposes you on the Salween. Your advance 
to the west cannot help but succeed. 

To take advantage of just such an opportunity, we have, during the past year, been equip- 
ping and training your Yoke Forces. If they are not to be used in the common cause our 
most strenuous and extensive efforts to fly in equipment and to furnish instructual personnel 
have not been justified. They should not be held back on the grounds that an amphibious 
operation against the South Burma coast is necessary prior to their advance. Present devel- 
opments negate such a requirement. The Jap has deployed the bulk of seven divisions in his 
operations on the Arakan, the Chindwin, and in the Mogaung Valley. 

I do hope you can act. 36 

In this radio the President did not refer to the question of the American 
military mission to north China, and the subject was dropped for the time 
being. The War Department believed it was far less important than persuad- 
ing the Generalissimo to send his armies across the Salween and to permit the 
Chinese Army in India to continue its advance down the Mogaung valley of 
north Burma. 37 

When the President's strong message arrived in Chungking, the General- 
issimo was ill, so Hearn gave it to Madame Chiang for delivery. At a tea party 
in honor of U.S. Army Day, Madame Chiang called General Hearn to one side 
and expressed her concern over the tone of the President's message. She feared 
that such a communication from Roosevelt would jeopardize rather than im- 
prove the chances of moving the Y-Force across the Salween River. As 
Madame Chiang continued, Hearn concluded that the message had not been 

56 (1) Stii 'well's Mission to China \Ch. VIII. \ (2) Rad WAR 17956, Roosevelt to General- 
issimo, 3 Apr 44. Item 2164, Bk 6, JW5S Personal File. Hearn relayed this message as Memorandum 
175 to the Generalissimo on 4 April 1944. 

57 (1) CM-OUT 25588, Marshall to Stilwell, 20 Apr 44. (2) The President renewed his 
proposals for a mission to north China during Vice-President Henry A. Wallace's mission in 
late June 1944. Se < j Chapter XJ below. 



delivered. He so reported to Marshall who relayed Hearn's impressions to 
Roosevelt. 38 The President immediately ordered that future messages from him 
to the Generalissimo were to be delivered in person to the Generalissimo by 
the senior U.S. Army officer present in Chungking. The English text was to be 
accompanied by a translation. 39 

Within a few days after the President had rejected the Generalissimo's prot- 
estations of 27 March 1944, fresh warnings of trouble in China were delivered, 
this time by Chennault to Stilwell. Where the Generalissimo had spoken in 
general terms, Chennault was specific. Sharing the general unawareness that 
the Japanese since the summer of 1943 had moved five good divisions from 
China to face the Allies elsewhere, and looking rather at the new and menac- 
ing Japanese concentrations to the north and south of Honan Province, Chen- 
nault put his warnings in the dramatic phrase that the disposition of the Jap- 
anese ground forces was the most menacing since Pearl Harbor. He believed 
that they had planned two offensives, one to overrun the Honanese link of the 
Peiping-Hankow railroad, the other to take Changsha. "In considering the 
likelihood of effective resistance to these anticipated offensives," wrote Chen- 
nault, "I need not point out to you the underlying weakness of the Chinese 
Armies." Chennault did not believe he could meet a determined Japanese air 
offensive and still give proper air support to the Chinese Armies. 

General Chennault argued that his problems were basically those of logis- 
tics. His supply level in east China was but 40 percent of that necessary, and 
the eastern line of communications could deliver but 2,500 tons a month. 
Chennault suggested the following steps: (1) deliver 8,000 tons of aviation 
supplies to China in May 1944 and build up thereafter to 10,000 in July; (2) 
give fullest support to improvement of the eastern line of communications; 
(3) establish first reserves of 10 percent of total air strength in China and an 
equal amount in India. If these steps, or a massive diversion of supply tonnage 
from the B-29's, could not be taken, then there was nothing left but to tell 
Washington the security of the China base was in doubt, and to tell the Gen- 
eralissimo he could not have effective air cover for his armies. 40 

38 Rad CFB 15828, Hearn to Stilwell, 7 Apr 44; Rad CFB 15835, Mme. Chiang to Stilwell, 
8 Apr 44; Rad CFB 15917, Hearn to Marshall, 10 Apr 44. Items 2179, 2181, 2194, Bk 6, 
JWS Personal File. 

39 (1) Memo, Roosevelt for Marshall, 3 May 44; Memo, Gen McNarney for Roosevelt, 8 
May 44, sub: Delivery of Msgs to Generalissimo from President. Item 58, Folder 1, General- 
issimo Chiang Kai-shek (GMO CKS), OPD Exec 10. (2) CM-OUT 33493, Marshall to Stilwell, 
7 May 44. 

40 (1) Rad CAK 456, Chennault to Stilwell, 6 Apr 44. Item 2174, Bk 6, JWS Personal 
File. (2) Ltr, Chennault to Stilwell, 8 Apr 44; Rad CAK 595, Chennault to Stilwell, 10 Apr 
44. Ltr, Chennault to Wedemeyer, 6 Jul 45 (hereafter, The Chennault- Wedemeyer Letter), 
Items 25, 26. WDCSA 091 China, 15 Aug 45. (3) Gen. Yasuji Okamura differed with Chennault's 
estimate of the Japanese situation on the ground that between summer 1943 and April 1944 
the Japanese transferred four good divisions from China to Pacific areas and replaced them only 
with raw brigades. Okamura apparently forgot the 15th Division, which left China for Burma 
in summer 1943. Japanese Officers' Comments, Incl 3, Okamura. (4) Japanese Study 129, map 
facing page 30, shows the concentrations that Chennault referred to. Page 11 of the monograph 
lists Japanese troop movements in China. 



Chennault's requests reached Stilwell at an awkward time. The Japanese 
were placing heavy pressure on India, and Mountbatten, who was Stilwell's 
superior, had diverted twenty transports from the Hump only three weeks 
earlier. If the Japanese went on to Assam, all would be over in CBI. In his 
reply, therefore, Stilwell stated that until the Imphal situation improved he 
saw no possibility of increasing the flow of supplies to Chennault. In Stilwell's 
opinion, the Combined Chiefs of Staff would not give priority to meeting an 
anticipated Japanese offensive in China when there was an actual and current 
crisis in India. "You will simply have to cut down on activity to the point 
where you can be sure of reasonable reserves for an emergency." Chennault's 
8 April letter did not reach Stilwell until about 21 April. Meanwhile on 12 
April, Stilwell warned Chennault (presumably because of the current attempts 
by the President to persuade the Generalissimo to commit the Y-Force) not 
to send a gloomy estimate of the situation to the Chinese leader. 41 

The Chinese Decide To Cross the Salween 

No formal reply was ever sent by the Generalissimo to the President's final 
demand that China attack. However, over the next ten days, the members of 
the National Military Council argued the merits of crossing the Salween River 
among themselves and with General Hearn. Hearn ordered Dorn, chief of staff 
of the American personnel working with the Y-Force, to come to Chungking 
to present the latest and most complete data on Y-Force's state. Dorn in turn 
persuaded Gen. Wei Li-huang to include Gen. Hsiao I-hsu, chief of staff of the 
Chinese Expeditionary Force, in a Chinese deputation. 

During the first few days of the discussions it seemed obvious to the Amer- 
icans that even the Chinese senior officers closest to the Generalissimo were 
not familiar with the contents of the President's 4 April message. However, the 
gist of a subsequent message of 10 April from Marshall to Hearn was given 
to the Chinese staff. Making a logical extension of the President's views, the 
radio from Marshall told CBI Theater authorities that unless the Y-Force 
moved, lend-lease shipments for it should end. Stilwell instructed Hearn to 
comply at once. Accordingly, Hearn told the War Department that he pro- 
posed to transfer the Y-Force's April allocation of Hump tonnage, 734 tons, 
to the Fourteenth Air Force. Furthermore, he planned to cancel the China Na- 
tional Aviation Corporation's contract to fly lend-lease to China and return the 
corporation's lend-lease aircraft to the Air Transport Command. 42 

41 (1) See |Ch. V J above. (2) Rad SH 7, Stilwell to Chennault, 9 Apr 44. Item 112, Bk 6A, 
JWS Personal file. (3) Rad SH 18, Stilwell to Chennault, 12 Apr 44. SNF 31. (4) Rad SH 
19, Stilwell to Chennault, 12 Apr 44. Item 2205, Bk 6, JWS Personal File. Chennault received 
Stilwell's message on 13 April 1944. 

42 (1) CM-OUT 20146, Marshall to Stilwell, 7 Apr 44. (2) Rad CRA 966, Sultan to Hearn, 8 
Apr 44; Rad CFB 15917, Hearn to Marshall, 10 Apr 44; Rad CHC 192, Stilwell to Hearn, 11 
Apr 44; Rad CFB 15985, Stilwell to Marshall, 11 Apr 44; Rad CFB 15986, Hearn to Chennault, 
11 Apr 44; Rad CFB 16010, Hearn to Stilwell, 12 Apr 44; Rad CFB 16029, Hearn to Stilwell, 12 
Apr 44. Items 2185, 2194, 2195, 2196, 2197, 2201, 2203, Bk 6, JWS Personal File. 



Word of Hearn's intentions soon reached Gens. Ho Ying-chin, Lin Wei, 
and Liu Fei, whom the members of Stilwell's staff had long believed to be op- 
posed to the Burma campaign. Meanwhile, General Hsiao, who realized the 
implications of Hearn's intended action, was urging the principal Chinese mili- 
tary personages, including the Generalissimo, to order an advance by Y-Force. 
On 12 April, Gen. Lin Wei asked Hearn to hold back orders on the tonnage 
cut for Y-Force because he was sure that there would be "positive action" on 
the use of Wei's troops within forty-eight hours. Hearn of course agreed to 
wait for two days, but, considering that the Chinese tended to vacillate and 
delay, he meanwhile insisted that the Chinese agree among themselves on a 
definite plan of attack, a command arrangement, and a date. 43 

On the evening of 12 April, Dorn held a dinner party for the various Chi- 
nese dignitaries present for these exchanges over the Y-Force. He may well 
have had in mind the ancient practice of the Persian general staff, which, it has 
been said, offered the advantage of combining the impetuosity of youth, the 
genius of experience, and the caution of maturity by the simple device of con- 
sidering a plan in all stages of intoxication and sobriety. As Dorn and Hsiao 
moved about the room, the trend of the guests' comments was all toward an 
offensive. At the end of the evening, General Ho was willing to confer with 
the Generalissimo about the matter. Next morning, Ho was still sure that 
matters could be arranged. 

True to his promise, General Ho, as Minister of War and Chief of Staff, 
gave formal approval to the Salween crossing on 14 April. In accord with Chi- 
nese custom, he gave the word before witnesses and made it binding by 
stamping his chop on the order. But before he placed the irrevocable seal in 
place, Ho made what Dorn thought a final attempt to block the operation. 

At the last moment, Ho asked Dorn for his personal assurance that the 
American Y-Force Operations Staff (Y-FOS) would assume four responsibil- 
ities during the early phase of the campaign: (1) to ferry 50,000 Chinese troops 
across the Salween; (2) to give air support; (3) to co-ordinate the American- 
trained artillery battalions; (4) to share Wei Li-huang's command responsibil- 
ity of feeding and supplying munitions to the Chinese Expeditionary Force. 
Dorn was intimately aware of the neglected state of Wei's forces, and he be- 
lieved that he would have to act for Stilwell in agreeing. Dorn surmised that 
Ho expected him, a junior staff officer, to decline, thereby excusing the Chinese 
from the operation. Ho pressed his seal upon the paper; the operation was for- 
mally agreed to. 44 

Having promised to attack the 56th Division, General Ho hastened to assume 
for China full responsibility for the decision to cross the Salween River. Ad- 

4 ^ ( 1) Rad CFB 16029, Hearn to Chennault and Stilwell, 12 Apr 44; Rad SH 23, Stilwell to 
Hearn, 13 Apr 44; Rad 16021, Hearn to Stilwell, 13 Apr 44. Items 2203, 2209, 2210, Bk 6, JWS 
Personal File. (2) Stilwell Diary, 12 Apr 44. 

44 ( 1) Ltr, Dorn to Stilwell, 16 Apr 44. SNF 35. (2) Interv with Dorn, May 48. 



dressed to Marshall, his radio of 14 April was understood by the Americans to 
be China's reply to the President's 4 April message: 

China has been working on plans for offensive action against Japanese with full inten- 
tion to put them into effect, using troops which have not received American equipment as 
well as Yoke Force. China has always realized her position with regard to offensives by 
United Nations, and it has only been because of time and lack of essential equipment that 
such action has not taken place before this time. You can rely on China doing her share, but 
it is hoped that you understand her difficulties. Decision to move part of Yoke Force across 
Salween was made on initiative of Chinese without influence of outside pressure, and was 
based on realization that China must contribute its share to common war effort. 

Marshall acknowledged Ho's pledge, and promised to relay the good news to 
Roosevelt at Warm Springs, Georgia. Meanwhile, Hearn restored Y-Force's 
full tonnage allocations. 45 

Chennault Renews His Warnings 

Immediately after the Chinese agreed to drive across the Salween toward 
Burma, Chennault again warned that the Japanese were about to move in east 
China, and by implication urged cancellation of the Salween offensive. On 15 
April, at the Generalissimo's request, Chennault as Chief of Staff, Chinese Air 
Force, submitted to the Generalissimo an estimate of the situation with respect 
to the air war in China. Chennault's air estimate was very similar to his 8 April 
letter to Stilwell, but with some significant additions. It will be recalled that 
on 12 April Stilwell had ordered Chennault not to submit such a paper to the 
Chinese leader. 46 

General Chennault wrote to the Generalissimo: "It is unnecessary to point 
out that all the new military equipment brought into China in the past two 
years has been assigned to the Chinese Armies on the Salween Front. Both 
equipment and many tens of thousands of troops have actually been borrowed 
for the Salween front from the Chinese forces which must meet the enemy of- 
fensives in Central and East China." In his next paragraph Chennault con- 
tinued "In addition to resisting the two enemy offensives [which he was pre- 
dicting], it is desired that the Chinese ground forces assume the offensive 
themselves, with the objective of taking Ichang, Shasi, and perhaps Han- 
kow." 47 Chennault therefore in mid-April 1944 set a very high rating on the 
capabilities of the Chinese forces, for the Japanese held Hankow with eight 
divisions. A Chinese attack on eight Japanese divisions was of a different order 
of magnitude from one across the Salween on the 56th Division. 

After listing for the Generalissimo the logistic difficulties that hobbled the 
Fourteenth Air Force, Chennault closed with the warning: 

45 Rad CFB 16100, Ho for Hearn to Marshall, 14 Apr 44; Rad CFB 16145, Hearn to Marshall, 
15 Apr 44; Rad WAR 23478, Marshall to Ho, 15 Apr 44; Rad CFB 16384, Ho for Hearn to 
Marshall, 21 Apr 44. Items 2214, 2220, 2225, 2250, Bk 6, JWS Personal File. 

46 (1) Lti, Ferris to Chennault, 8 May 44, with 1st Ind, Chennault to Ferris. 14 May 44, 
sub: Info for Theater Comdr. SNF 31. (2) Stilwell's order is radio cited in |note 41 (4). | 

47 Memo, Chennault for Generalissimo, 15 Apr 44, par. 4. SNF 31. 



Under the circumstances, therefore, it is necessary to inform your Excellency that the 
combined air forces in China, excluding the VLR [B-29] project, may not be able to with- 
stand the expected Japanese air offensive and will certainly be unable to afford air support to 
the Chinese ground forces over the areas and on the scale desired. In order to put the air 
forces on a footing to accomplish these missions, drastic measures to provide them with ade- 
quate supplies and adequate strength must be taken. As the Japanese threat appears to be 
immediate, such measures should be taken without further delay. 48 

On completing his study, Chennault tried to give a copy to General Hearn. 
Hearn was ill, so without further delay, Chennault presented his estimate to 
the Generalissimo. The following day a copy was left with General Ferris, who 
was acting as chief of staff vice Hearn. Circulated among StilwelPs principal 
staff officers in Chungking, Chennault's estimate drew emphatic dissents. G-2 
stated that while officially the Chinese professed great fear of an offensive by 
the Japanese, Chinese intelligence saw no indications and was unconcerned. 
G-3 rejected Chennault's statement that the Chinese divisions in Yunnan had 
been borrowed from other fronts in China, observing: "The implication of this 
paragraph is that China will fall because the ground forces are employed in the 
wrong place." 49 Apparently, Chennault's 15 April estimate was about three 
weeks in coming to Stilwell's attention. 

At this time, informational copies of the 8 and 10 April 1944 exchanges be- 
tween Chennault and Stilwell were circulating among interested headquarters 
all the way up to General Arnold of the AAF in Washington. Throughout 
April, and indeed later, Chennault believed his warnings were being ignored. 
Actually they caused serious concern all along the chain of command. General 
Stratemeyer in India was alarmed because final preparations were being made 
for reception of the B-29's in China. Matterhorn would be useless if Strate- 
meyer and his staff read Chennault's radios correctly. Other echelons reacted 
similarly. Headquarters, CBI Theater, considered curtailing Chennault's oper- 
ations so that there might be ample fighter cover for Matterhorn against 
the Japanese air offensive Chennault was predicting. Arnold suggested that the 
Fourteenth Air Force receive the major share of all Hump tonnage. Operations 
Division was so alarmed that it persuaded the AAF to delay Matterhorn 
until Stilwell was sure the B-29's could be safely based in China. Stratemeyer 
immediately queried Chennault on this point and suggested a conference at 
Dinjan, India, to cover the entire matter with Stilwell. 50 

On 16 April Chennault declined to meet Stilwell and Stratemeyer. He as- 

48 Ibid. . . 

"9(1) Ind citec jn. 46(1 % (2) The comments of Stilwell's Chungking staff are on the covering 
slip with which Ferris circulated Chennault's estimate to G-2, G-3, G-4, and the air officer. 
SNF 31. _ 

,0 (1) Ltr and Rad citec jn. 40~"| (2) Rad CAK 879, Chennault to Stratemeyer, 16 Apr 
44; Rad SH 58, Stilwell to CrTeTmautt; 21 Apr 44; Memo, Stratemeyer for Chennault, 20 Apr 
44, sub: VLR Opns from China; Rad CAK 1215, Glenn to Chennault, 24 Apr 44; Rad CAK 
1284, Chennault to Stilwell, 26 Apr 44; Rad CHC 1016, Stilwell to Chennault, 30 Apr 44. The 
Chennault-Wedemeyer Letter, Items 28, 29, 30, 31, 32, 33- (3) Memo with Incls, Arnold for 
Handy, 17 May 44, sub: Opns of U.S. AAF Units from China. Case 375, OPD 381, A47-30. 



sured General Stratemeyer that he could defend the Cheng-tu B-29 base. 
Claiming that his 8 April letter had been misunderstood, Chennault explained 
his "intentions were to acquaint and point out the increased danger of hostile 
air attack if Peking-Hankow Railroad is held by Japs." Chennault, therefore, 
saw grave danger, but now appraised it as "increased danger of hostile air 
attack." 51 

Operation ICHIGO 

The planning for Japan's 1944 offensive in China, an operation whose open- 
ing phases would form an ominous background to the Chinese offensive across 
the Salween, began in the fall of 1943. Japanese staff studies then weighed the 
advantages to be gained by taking the east China airfields and preventing air 
raids on Japan, the disruption of Chinese activities in south China, the open- 
ing of secure land communications to Southern Army via Indochina, and the 
overthrow of the Nationalist regime. Further study focused on one objective 
that would satisfy the concern felt by Imperial General Headquarters about the 
menace of the B-29's, and meet the situation that faced China Expeditionary 
Army as a result of the Fourteenth Air Force's attacks on its lines of communi- 
cations. This objective was the east China airfields, the current base for Chen- 
nault's tactical aircraft, and a potential base for the B-29's. 52 \ (Map 18*) 

In the later opinion of General Okamura, who in January 1944 commanded 
North China Area Army and subsequently assumed command of all the Jap- 
anese forces in China, the "extremely effective" attacks of the Fourteenth Air 
Force made it very difficult to supply the Japanese forces in the Hankow area. 
Because the Japanese air force was unable to check Chennault's airmen, the 
Japanese had no alternative "but to plan to destroy Chennault's bases of 
operation from the ground." 53 

On 17 January 1944 Imperial General Headquarters resolved to order China 
Expeditionary Army, Gen. Shunroku Hata commanding, to seize the east China 
airfields, and take the Hunan-Kwangsi, Canton-Hankow, and Peiping-Han- 
kow railways. After the war, General Okamura recalled that the Peiping- 
Hankow line had to be taken over by the Japanese because the Fourteenth Air 
Force attacks on Japanese river shipping along the Yangtze River were making 
that line of communications "extremely unsafe." 

ICHIGO, therefore, had its limits. It did not at any time aim at taking 
China's capital, Chungking, or the Hump terminal, Kunming. Nor was there 
any co-ordination between ICHIGO and the Japanese drive on Imphal. Look- 
ing back on the war years, Hata thought it unfortunate that Imperial General 

51 Rad CAK 879, Chennault to Stilwell and Stratemeyer, 16 Apr 44. The Chennault-Wedemeyer 
Letter, Item 2£__- 

52 (1) Se JCh. IJ above. (2) Japanese Study 45, p. 109- 

53 JapaneseCHncers' Comments, Incl 3, Okamura. 

*For an explanation of enemy unit symbols, see note) 82 (3), page 42] above. 



Headquarters had given no instructions about co-ordination or co-operation be- 
tween the two operations. Hata's staff paid little attention to StilwelPs oper- 
ations in north Burma and did not consider that the Japanese offensive against 
Imphal concerned them: 

That is, it believed that the operational objectives of the U.S. forces were to reopen the 
North Burma land route to support Chiang Kai-shek's forces and to check Japanese forces' 
advance on India and it thought very little of them. It did not have the acute perception to 
see that the U.S. forces operations in Burma were a part of the India-Burma-China-Pacific 
Operation. 54 

The reinforcements that Imperial General Headquarters sent to China seem 
to have been planned to provide artillery and service troops for Hata's divi- 
sions, most of which had little of either and so were ill adapted to active cam- 
paigning. From Manchuria there came one division, the 21th, four battalions of 
field artillery, four battalions of mobile antiaircraft artillery, two mortar bat- 
talions, three battalions of engineers, four field replacement units, six animal 
transport companies, and fifteen motor transport companies. Three air regi- 
ments were also transferred. These were all to return to Manchuria by the be- 
ginning of 1945. From Japan Imperial General Headquarters sent one independ- 
ent infantry brigade, two battalions of mobile antiaircraft artillery, two com- 
panies of antiaircraft machine guns, four field replacement units, plus signal, 
road construction, transport, and medical units. In China Hata organized two 
regiments of engineers, two battalions of artillery, and signal units. 

Since a number of Hata's divisions were organized as Class C units, and 
therefore had no artillery, Imperial General Headquarters intended to raise some 
to B standard, with six batteries of artillery, and others to A, with three bat- 
talions. 55 It may be assumed the artillery reinforcements were for that end. But 
since the 32d and 35th Divisions were to be sent out of China in the spring of 
1944, obviously Hata was not going to receive major reinforcements for 
ICHIGO. And, after the completion of ICHIGO, two more divisions were to 
be sent from China to Manchuria, and three from China to French Indochina. 56 

Imperial General Headquarters also ordered Hata to make maximum use of 
local resources. All that was sent to China to augment Hata's supplies were 
10,560,000 gallons of gasoline for trucks and 2,640,000 gallons of aviation gaso- 
line, plus enough ammunition, as a postwar Japanese account rather vaguely 
put it, to sustain four divisions in one engagement. In 1943 Chennault and his 
adherents had argued that the Japanese could find troops to overrun the east 
China airfields only by withdrawing garrisons from the Pacific and thereby 

,4 (1) Japanese Studies 78, 129. (2) Stilwell's Mission to Ckin dj Ch. IXl| (3) IGH Army Order 
921, 17 Jan 44. (4) Japanese Officers' Comments, Incls 2, 3, Hata, Ukamura. 

" (1) Japanese Study 129, p. 22. (2) Contemporary U.S. divisions all had thirty-six 105-mm. 
howitzers and twelve 155-mm. howitzers as an organic part of the division. This was regarded as 
the absolute minimum for operations against a weak adversary. Normal practice, on entering 
combat, was to attach added battalions of artillery. 

,6 Japanese Study 29, p. 22. 



sacrificing their conquests. But when the time came for their major effort in 
China, the Japanese continued their policy of transferring veteran divisions 
from China to face the Allies elsewhere. 57 

Beginning in early February 1944 the significant Japanese redeployments 
for ICHIGO were made south of the Great Wall, within Hata's command. 
Japanese units from north China moved south and east, leaving vacuums into 
which it seems reasonable to assume Communist guerrilla units moved. Thus, 
the Mongolia Garrison Army (Lt. Gen. Yoshio Kozuki) lost the 3d Tank Divi- 
sion. The North China Area Army (General Okamura) reassigned the 110th 
Division. In all, 7 of the 23 divisions of the Japanese military establishment in 
China were involved in such shifts. On the eve of its last offensive, China Ex- 
peditionary Army had 24 divisions, 12 independent mixed brigades, and 14 in- 
dependent infantry brigades, plus 230 aircraft of the newly activated 5th Air 
Army (Lt. Gen. Takuma Shimoyama). Total personnel numbered approxi- 
mately 820,000 men. 58 From the above, fifteen divisions, four independent 
mixed brigades, and one independent infantry brigade were ultimately to take 
part in the ICHIGO operation. 59 

General Hata divided ICHIGO into two major parts. The first, K0G0, was 
a preliminary to his major effort, TOGO. K0G0 was to clear the Chinese off 
the railway lines that ran north to the Yellow River from Hankow. For the 
operation, North China Area Army, to drive south across the Yellow River, 
would use the 12th Army (one armored and three infantry divisions, one in- 
dependent mixed brigade, one cavalry brigade, and one independent infantry 
brigade), under Lt. Gen. Eitaro Uchiyama, and would feint toward the west 
with some elements of the 1st Army, commanded by Lt. Gen. Teiichi Yoshi- 
moto. About six weeks were allowed for the operation. 

After K0G0 had scattered the Chinese forces between the Yangtze and the 
Yellow River, "particularly the Nationalist," cautioned Imperial General Head- 
quarters, and had cleared the rail lines from Peiping to Hankow, TOGO would 
unfold in three phases. Capture of Heng-yang was to be first. Kweilin and 
Liuchow were to be taken in the second. Phase three would include capturing 
Nan-ning, opening the Canton-Hankow railroad, and overrunning the Four- 
teenth Air Force's fields at Sui-chuan and Nan-hsiung. For TOGO, 11th Army 
(Hankow) would have nine divisions, the 23d Army (Canton), under Lt. Gen. 
Hisakazu Tanaka, would have two divisions for its drive north to meet the 11th 
Army, and one or two divisions would be in central reserve. TOGO was to 
begin in the summer and be completed in five months. 60 

To protect the great Japanese supply center at Hankow, General Hata 

57 Bks VII, IX, Hopkins Papers. (2) Japanese Studies 78, 82, 129. 

58 (1) Japanese Studies 78, 82, 129. (2) Japanese Officers' Comments, pp. 23-24. (3) IGH 
Army Order 928, 1 Feb 44; Order 933, 5 Feb 44; Order 945, 15 Feb 44. (4) Ltr, Murphy to Ward, 
18 Nov 52. OCMH. 

"Japanese Officers' Comments. 

60 (1) Japanese Studies 78, 129- (2) IGH Army Dirs 1810, / Jan 44; 1830, 1 Feb 44. 



formed the Wuchang-Hankow Defense Army, and to it allotted the 39th Division, 
one independent mixed brigade, four independent infantry brigades, and three 
field replacement units. Up the river from Hankow, in order to protect Chung- 
king from a drive up the Yangtze, the Generalissimo had the V and VI War 
Areas with forces that in October 1943 the Japanese had estimated at eleven 
armies. The size of the garrison the Japanese left at Hankow suggests a high 
degree of confidence that these eleven Chinese armies would not attack Han- 
kow while ICHIGO was under way. 61 

East China's Defenders on the Eve of ICHIGO 

The exact strength and quality of the Chinese forces that faced Hata were 
unknown to either Stilwell's or Chennault's headquarters. 62 In Hata's path lay 
the IX War Area (Gen. Hsueh Yueh), and the IV War Area (Gen. Chang 
Fa-kwei). In Kweilin was the headquarters of Li Chi-shen (who bore the hon- 
orary title of marshal). He held powers of command over the east China war 
areas that the Americans believed to be nominal and purely honorary, though 
they regarded him as a powerful politician. Kweilin was also important as the 
headquarters of Gen. Pai Chung-hsi, who in the Nationalist Government 
hierarchy held the post of head of the Board of Training. He was important in 
his own right in that he commanded the loyalty of a group of divisions in east 
China whose exact strength was unknown but whose ability was highly rated 
by Americans and Chinese both. After the war the Japanese wrote they had ex- 
pected to find the Changsha area held by 13 to 14 Chinese armies totaling per- 
haps forty divisions, and that about 20 Chinese armies numbering perhaps fifty- 
five divisions had been committed to the defense of Heng-yang. 63 

These forces had been little improved by the Zebra Force project, Stilwell's 
effort to build thirty good divisions in east China. As graduates of the Kweilin 
Infantry Training Center had begun to rejoin their units, Stilwell had wanted 
to capitalize on their training. By organizing Americans into traveling instruc- 
tional groups and sending them into the field he hoped to create the begin- 
nings of a liaison system such as that with the Y-Force on the Sal ween. 

Accordingly, on 1 January 1944, a Z-Force Operations Staff, modeled on 
the Y-Force Operations Staff, was organized. As he had with the Y-FOS Amer- 
ican staff, Stilwell kept command of the little group of Americans and named 
Brig. Gen. Malcolm F. Lindsey as chief of staff. 64 

61 Japanese Studies 77,-Zfi 17 9 130. 

62 See notes ^("M ~- ^o*f i\\ 

63 (1) See sect-ion "The isast China Crisis Grows," Chapter XI, below. (2) Japanese Study 78. 

64 (1) Se dCh. I.l above. l'2\ History or Z-FUb. 1 Ian- 41 Uct 1944. (Hereafter, History of 
Z-FOS.) OCMH™(3) Stilwell's keeping command of the Y-Force and Z-Force Operations 
Staffs was interpreted by Dorn as largely motivated by personnel problems within the American 
organization in CBI, not by any plans of personal advancement in China Theater. Moreover, 
correspondence from Dorn and Lindsey to the Chinese would then be by Brigadier Generals 
Dorn and Lindsey on behalf of Lieutenant General Stilwell so that the relatively junior Dorn 
and Lindsey would not be as it were directly addressing very senior Chinese officers. 



General Lindsey set up a headquarters in Chungking, assembled his aides, 
and began work. Discussions with General Ho and the Chinese General Staff 
soon suggested to Lindsey and the Americans that the War Department deci- 
sion to limit Zebra Force to 10 percent of its full complement of lend-lease 
had had an adverse effect on the Chinese. As the weeks went by, Lindsey and 
his staff received the very strong impression that so far as the Chinese were 
concerned lend-lease in quantity must precede any action on their part. The 
Chinese would not set up an over-all command structure for the Zebra divi- 
sions, and American hopes for an improved Chinese force in east China rapidly 
faded. 65 

After weeks of conferring, Lindsey's G-4, Col. Frederic W. Boye, on 26 
March 1944 in effect urged scrapping the plan to build a second thirty divisions 
and accepting instead a Chinese suggestion: 

1. As has been and as will be brought out at G-4 conference, equipment for both Chinese 
and Americans will be meager. ... It is to be noted however that we are getting some 
equipment which can be put to immediate use. . . . 

3. Personnel ordered to the [CBI] theater for Z Force has been cut on two occasions and 
is being side-tracked on all sides with the approval of the Theater Commander [Stilwell was 
using Z-Force personnel in Burma since the Chinese were not offering a great deal of co- 
operation to Z-Force]. In effect, therefore, we do not have sufficient for our initial essential 
group for our second 30 divisional installation. . . . 

5. The National Military Council has designated 30 divisions which are to comprise the 
Z Force. [Boye's marginal notes located these as eight in the Kweilin area, six along the 
Yangtze, and 16 near Hsian.] These divisions are so widely separated and so distant from 
possible supply bases that it would be impossible to maintain supply and communications 
to them under the limitations in this [China] theater. . . . 

6. The Chinese are reluctant to have Americans go empty-handed to their units to take 
over any training without equipment. If the equipment which will be available within the 
next year [under the War Department's 10 percent policy] is divided into 30 parts each such 
part will be practically zero. My information leads me to believe' that in many divisions there 
is no training going on at the present time. This fact the Chinese wish to conceal and it is 
for that reason that they do not wish the Americans there. 

7. If we concentrated our Chinese equipment in the hands of a smaller number of units 
we would have something to work on with this fewer number. Likewise, the problem of turn- 
ing equipment over to the Chinese would be solved in that we would retain American con- 
trol along established routes to delivery to Army Headquarters. The Depot facilities at 
Kweilin are adequate to handle the storage and distribution involved. . . . 


It is my conclusion and recommendation that we accept the Chinese proposal on the 
concentration of equipment now available and that we secure approval of the Theater Com- 
mander toward concentrating our immediate installations and training attentions to the four 
Armies generally located at Luichow, Kukong, Hengyang, and Changsha; that as soon as 
the basic essential equipment is in our hands we dispatch Army and Division teams by rail 
to those places and otherwise proceed with our training plan; that until other equipment is 

65 (1) History of Z-FOS. (2) Z-Force Journal. KCRC. 



forthcoming we supply no equipment to the other Armies and divisions in the river valley 
areas and to those further to the north. 66 

The pressure of events eventually began to force American plans into the shape 
Boye had recommended. 

With the arrival in east China by the end of May 1944 of a very meager 
amount of arms for Zebra, the Generalissimo sent Lt. Gen. Lo Cho-ying to 
Kweilin to accept delivery. General Lo acted as a representative of the National 
Military Council and was in no sense a commander for the divisions the Na- 
tional Military Council had earlier named. As the Japanese were moving to 
their assembly areas, Lindsey in mid- April began preparing to move to Kwei- 
lin. He probably hoped that since a trickle of lend-lease was actually beginning 
to flow to east China, Lo might arrange for Lindsey to send U.S. instructional 
teams to the divisions designated by the Chinese for the Zebra project. 67 

On the night of 17 April, the Japanese 37 th Division crossed the Yellow 
River. Was it one more training expedition, or a rice raid? That day, Head- 
quarters, CBI Theater, radioed. G-2 in Washington that recently 239 Japanese 
troop trains had been reported as passing through Hsinhsiang. Stilwell's staff 
also reported that the Japanese were building their independent mixed brigades 
into full-scale divisions, that there were ominous troop movements around 
Hankow. 68 On 19 April, the Japanese 110th and 62d Divisions began moving 
south along the Peiping-Hankow railway, and the KOGO phase of ICHIGO 
was under way. 69 

Initial Reactions to ICHIGO 

While the Japanese waited in their assembly areas along the Yellow River, 
Stilwell and Stratemeyer had been discussing the problem presented by 
Chennault's warnings of a Japanese air offensive. Obviously, Stilwell took 
Chennault's warnings at face value, for on 17 April Stilwell directed that 
Chennault's primary mission be defense of the B-29 fields at Cheng-tu "even 
at the expense of shipping strikes and support of the Chinese ground forces, 
dependent upon Japanese reaction to operations from the Chengtu area." The 
order reached Chennault about 26 April. 70 

Since the Japanese offensive was now under way, and was emphatically a 
ground offensive, Chennault was startled by a directive that tied him to defense 
of a remote area in west China. A strong force of fighters was based between 

66 Memo, Boye for Lindsey, 26 Mar 44, sub: Observations on Z-Force Problems. Boye's 
italics. OCMH. 

67 (1) History of Z-FOS. (2) Recordings of Diary of 1st Lt Dwight E. Brewer, Adjutant 
Gen, Z-FOS. OCMH. (3) Extracts from Col Boye's diary. OCMH. (4) CM-IN 8737, Stilwell 
to Marshall, 11 Jun 44. 

68 (1) Japanese Study 129. (2) CM-IN 13041, Hearn to MILID G-2 WDGS, 18 Apr 44. 

69 Japanese Studies 78, 129- 

70 (1) See pp. 314-16, above. (2) Memo, Stratemeyer for Chennault, 20 Apr 44, sub: VLR 
Opns From China. The Chennault- Wedemeyer Letter, Item 30. 



the Japanese and Cheng-tu. Weighing the problems of giving fighter cover to 
Cheng-tu against those of helping the Chinese to stop the Japanese drive and 
preparing to support the steadily approaching Chinese offensive on the Salween 
front, Chennault assured Stilwell that the defense of Cheng-tu was "child's 
play" in comparison and asked that Stilwell reconsider his directive. 71 
Stilwell replied: 

I am glad to hear that the defense of Chengtu is child's play. I had gathered from your 
letter of April eight that the security of China as a base for Matterhorn and other military 
operations against Japan might be in doubt. It is a relief to know that we have no problem 
at Chengtu and under these circumstances of course the question of action in emergency will 
not arise. Until it does, there is no intention of limiting the scope of your operations in any 
way." 2 

Therefore, Chennault would have operational freedom in meeting the 
Japanese threat. SOS in China went on with its plans to improve the eastern 
line of communications to Chennault's bases and approved Colonel Sheahan's 
recommendations. 73 Attempts were made to make up the April shortfalls in 
tonnage deliveries to the Fourteenth Air Force. However, Stilwell refused to 
declare that there was an emergency under which he, as U.S. theater com- 
mander, could divert the B-29 stockpiles to the Fourteenth Air Force as 
Chennault had requested on 8 April. Simultaneously, Roosevelt refused a bid 
from the Generalissimo for command of the B-29's, giving him instead purely 
honorific control as Supreme Commander, China Theater. 74 

The Japanese launched the KOGO phase of their offensive as the Chinese- 
American Composite Wing was establishing itself on the airfields at Hsian, 
En-shih, Liang-shan, and Nan-cheng. The wing comprised six fighter and one 
medium bomber (B-25) squadrons, and its mission was to protect the B-29 
fields at Cheng-tu, to destroy the Yellow River bridges in Japanese possession, 
and to neutralize the railway yards at Cheng-hsien and Kaifeng. The first mis- 
sion against the Yellow River bridges was therefore flown by the Fourteenth Air 
Force on 28 April by twenty-seven B-24's escorted by ten P-51's. They found 
the bridges hidden under a low ceiling, and the mission was not effective. On 
5 May a P-40 knocked out one span by dive bombing, but the Japanese quickly 
repaired it and supplemented the regular bridges with two ponton bridges 
across the silt flats of the river east of Kaifeng. Throughout the war the duel 

71 Rad CAK 1284, Chennault to Stilwell, 26 Apr 44. The Chennault- Wedemeyer Letter, 
Item 32. 

72 (1) Rad CHC 1016, Stilwell to Chennault, 30 Apr 44. The Chennault-Wedemeyer Letter, 
Item 33. (2) Chennault, Way of a Fighter, page 286, quoted Radio CHC 1016 but eliminated 
the passage: ". . . and under these circumstances of course the question of action in emergency 
will not arise. Until it does, there is no intention of limiting the scope of your operations in 
any way." 

73 (1) See pp. 1 292-931 above. (2) Bykofsky MS. 

74 (1) Rad CFB 16169, Hearn to Marshall, 15 Apr 44. Item 2224, Bk 6, JWS Personal File. 
(2) Rad cited n. 72(1). (3) Rad CFB 16433, Generalissimo to Roosevelt, 22 Apr 44. Item 
2253, Bk 6, JWS Personal File. 


BRIDGES ACROSS THE YELLOW RIVER were repeatedly bombed by planes of 
the Fourteenth Air Force. 

between the airmen and the Japanese antiaircraft gunners continued, with the 
Japanese successful in keeping the crossings open for operation. 75 

As April ended, defense of Honan Province seemed hopeless. The fear of an 
attack by the Japanese toward Hsian, gateway to Chungking and Cheng-tu, 
thoroughly alarmed the Generalissimo. His commanders along the Yellow 
River were ordered to prevent a juncture by the Japanese in Tung-kuan (East 
Gateway) Pass, whose capture would open the way to Hsian. In turn, Hsian's 
fall might blow up a storm of dissatisfaction among the Generalissimo's war 
area commanders south of the Yangtze, for many of them, the Americans 
reported, were now hinting of their dissatisfaction with his conduct of the war. 76 

At this moment, Stilwell reacted to Chennault's sending the air estimate of 
15 April to the Generalissimo. Considering that Chennault had acted in 
defiance of his explicit orders not to lay such a paper before the Generalissimo, 
Stilwell demanded an explanation in writing. 77 

" (1) Fourteenth AF History. (2) Japanese Study 82. 

76 (1) Japanese Study 78. (2) Rad 460, Lindsey to Stilwell, 13 May 44. Item 2534, Bk 7, 
JWS Personal File. (3) General Okamura stated after the war that the Japanese did not have 
the resources to attack Hs ian. Japanese Officers' Comments, Incl 3, Okamura. 

77 Ltr cite d] n. 46(l71 



While Chennault drafted his answer, the Generalissimo began applying 
pressure to bring more U.S. air power to bear against the Japanese. Calling 
General Ferris, then acting as chief of staff for Stilwell, to meet with him, the 
Generalissimo on 1 1 May asked that Stilwell order the fighter aircraft stationed 
at Cheng-tu to protect the B-29's in moving to Nan-cheng to support the 
Chinese forces in and around Loyang. He also asked that 500 tons of gasoline 
from the B-29 stores be diverted to support the fighters. The Generalissimo 
told Ferris that the next two weeks were vital, that he wanted the Fourteenth 
Air Force and his armies in close co-operation to try to stop the Japanese. 78 

Promptly transmitted to Stilwell by Ferris, the Generalissimo's request 
placed before Stilwell the problem of deciding when conditions in China 
Theater would reach a point that would force him to divert supplies from the 
B-29 project, Matterhorn, to which the President attached such importance. 
Stilwell's remark at Trident that "air coverage over nothing is in my opinion 
of little value" suggests he now hesitated to divert supplies from Matterhorn 
for a tactical air effort because he expected it to be futile in view of the low 
quality of the troops for whom it was exerted. He may also have recalled 
Chennault's statement that the greatest danger was from a Japanese air 
offensive, which would surely strike at the menace to the Japanese homeland 
which the B-29's presented. But other measures Stilwell could approve. He 
approved Stratemeyer's suggestion that the B-24's of the 308th Bombardment 
Group haul gas and oil to Chennault, and permitted Chennault to use the 
Cheng-tu P-47's. Feeling little sympathy for the Generalissimo in a predica- 
ment that Stilwell believed the Chinese leader had brought on himself by 
failing to reorganize his army, he thanked Stratemeyer for the suggestion, then 
added: "We must remember that he [the Generalissimo] has been assured by 
experts that air power can do the trick, and now he craves to see it done." 79 

On 12 May a radio from Chennault presented the Fourteenth Air Force's 
reasons for its inability so far to stop the Japanese. Chennault told Stilwell what 
he had been able to do to meet the Generalissimo's demands for more air sup- 
port. Three hundred and seventy tons of gasoline had been given to the Chinese 
Air Force and twelve P-51's had been sent to Liang-shan to reinforce the 
fighters there. Lack of airfields, limited supplies, dust, and poor communica- 
tions made it difficult for the Fourteenth Air Force to operate in the Loyang- 
Cheng-hsien area. Chennault concluded: "lack of tonnage for aviation supplies, 
and a general disbelief in Japanese offensive plans" handicapped him in his 
efforts to prepare for the Honan campaign. 80 

Then Chennault's reply to Stilwell's demand for an explanation arrived. 

78 Memo, Mme Chiang for Stilwell, 5 May 44; Rad CFB 17237, Ferris to Stilwell, 11 May 
44. Items 2507, 2525, Bk 7, JWS Personal File. 

79 (1) Memo 197, Ferris for Generalissimo, 14 May 44. Item 2535, Bk 7, JWS Personal File. 
(2) Rad CHC 1054, Stilwell to Stratemeyer, 13 May 44. SNF 130. 

80 Rad CAK 1946, Chennault to Ferris for Stilwell, 12 May 44. Item 2531, Bk 7, JWS Personal 



In it Chennault pointed out that he was chief of staff of the Chinese Air Force, 
which post gave him direct access to the Generalissimo. He had tried to present 
the paper to Hearn before giving it to the Generalissimo but Hearn had been 
ill. Such conflicts as this were perhaps implicit in his dual status as Command- 
ing General, Fourteenth Air Force, and as Chief of Staff, Chinese Air Force, 
Chennault continued, but he had left a copy of the study with Ferris so that 
Stilwell would be informed. Weighing this answer, Stilwell found himself far 
from satisfied with it. He concluded that Chennault had been insubordinate 
and should be relieved. 81 

Looking over Chennault's radios to him and the air estimate to the General- 
issimo, Stilwell concluded that Chennault was beginning to prepare a case to 
which he might appeal in years to come. Taking up his pen, Stilwell poured 
out the bitterness accumulated in the long feud that had so handicapped his 
efforts to prepare an effective Chinese Army, and wrote his analysis of 
Chennault's tactics: 

Chennault [stated]: The Chinese ground forces can protect the bases with the help of 
the 14th AF. . . . 

Chennault has assured the Generalissimo that air power is the answer. He has told him 
that if the 14th AF is supported, he can effectively prevent a Jap invasion. Now he realizes 
it can't be done, and he is trying to prepare an out for himself by claiming that with a 
little more [Stilwell's italics], which we won't give him, he can still do it. He tries to duck the 
consequences of having sold the wrong bill of goods, and put the blame on those who 
pointed out the danger long ago and tried to apply the remedy. 

He has failed to damage the Jap supply line. He has not caused any Jap withdrawals. On 
the contrary, our preparations have done exactly what I prophesied, i. e., drawn a Jap 
reaction, which he now acknowledges the ground forces can't handle, even with the total air 
support he asked for and got. 82 

The East China Army Written Off 

Soon after Stilwell angrily penned the analysis quoted above, Chennault 
on 18 May asked that a directive be issued to meet the increasingly grave situa- 
tion which threatened to involve loss of the east China bases, that adequate 
supply tonnage be given the Fourteenth Air Force, that CBI Theater head- 
quarters furnish information on what action it was taking to meet the threat, 
and that information be furnished as to just what would constitute an emer- 
gency justifying Stilwell in using the B-29's and their supplies within China 
Theater to stop the Japanese. 83 

81 (1) RadCFB 17552, Ferris to Stilwell, 19 May 44. Item 2542, B k 7. TWS Per sonal File. (2) 
Indorsement and Stilwell's handwritten notes attached to letter cited i nfnot^£^^1 

82 ( 1 ) The passage deleted for brevity's sake gives Stilwell's understanding that Chennault had 
promised 100 fighters, 35 bombers, and 5,000 tons a month of supplies would be enough for him 
to drive the Japanese from central China. The paper, handwritten in ink, was found in SNF 31, 
clipped to a radio dated 14 May 1944 and just before a letter from Fgm^dajed 11 May 1944. (2) 
Chennault, Way of a Fighter, p. 304. (3) Stilwell's Mission to China lG^- IX. | 

85 Ltr, Chennault to Stilwell, 18 May 44, sub: Estimate of Enemy Intentions in China. SNF 31. 



General Chennault supported his request for more supplies and a priority 
overriding that of the B-29's with an intelligence estimate by his staff that 
dismissed the Chinese Army as a factor in defending the east China airfields and 
by clear implication left the Fourteenth Air Force as the only Allied force that 
might save the situation. Estimating the strengths of the contending ground 
forces, Chennault's Assistant Chief of Staff, A-2, Col. Jesse C. Williams, set the 
Japanese strength in north China and in the Yellow River bend as eleven 
divisions, four independent mixed brigades, four infantry brigades, and one 
cavalry brigade. The Chinese, Colonel Williams went on, had not been willing 
to reveal their strength but no less than thirty-four divisions had been men- 
tioned in reports from the front, some of them units with a reputation of being 
excellent troops. 

Colonel Williams was not optimistic about the performance of the Chinese 
soldiers. He told Chennault: "Only at Szeshui [Ssu-Shui, Honan] have these 
troops offered advancing Japanese forces substantial resistance. Everywhere else 
Japanese columns have moved virtually at will. The Chinese have shown only 
slight evidence of either plan or capability to hamper Japanese movement or to 
regain lost territory." For the first time, he wrote, the Chinese were faced with 
Japanese divisions from Manchuria with plenty of mobile artillery and armor. 
This estimate, as has been noted, was in error, for General Hata had but one 
division (the 27th) from Manchuria. Because it was the first time since 
December 1941 that the Japanese had put forth a major effort in China, the 
contrast with their previous operations led Allied observers to explain it in 
terms of massive reinforcements from Manchuria. 84 

The speedy initial successes scored by ICHIGO, the poor combat perform- 
ance of the forces of the east China commanders, the acute supply problems of 
the Fourteenth Air Force, and the discord among the local American 
headquarters were a somber backdrop to the proposed Chinese effort along the 


After the conferences at Cairo and Tehran in December 1943, President 
Roosevelt's attitude toward China changed greatly. The Generalissimo's con- 
duct at Cairo, the Soviet promise to enter the war against Japan, the General- 
issimo's linking his request for a loan of $1,000,000,000 with the cancellation 
of the Andamans operation (Buccaneer), Chinese insistence on making the 
Americans literally pay to fight in China, the contrast between StilwelPs 
defeating the Japanese 18th Division with three Chinese divisions and the 

84 ( 1 ) As a measure of the increased effort that they put forth, the Japanese military expendi- 
tures in China jumped 700 percent hetween 1943 and 1944. USSBS, Effects of Strategic Bombing on 
Japan's War Economy, Table B-2, p. 85. (2) Memo, Williams for Chennault, 18 May 44, sub: 
Estimate of Japanese Capabilities on the China Front. The Chennault- Wedemeyer Letter, Item 
37 with Incl. 



Generalissimo's reluctance to engage the weakened 56th with twelve Chinese 
divisions, all played their part in the President's appraisal of Chiang Kai-shek 
as a soldier and as a statesman. The President's messages to the Generalissimo 
grew steadily harsher in tone, culminating in a threat to cut off lend-lease if the 
Generalissimo continued his refusal to attempt to break the blockade of China. 

While the President's attitude was changing, and as he leaned more toward 
support of Stilwell's views rather than the Generalissimo's, there were ominous 
Japanese stirrings in China. The Generalissimo as Supreme Commander, China 
Theater, had not organized an army able to meet eleven Japanese divisions on 
even terms, and Chennault's east China bases were now the object of a major 
Japanese effort. Unfortunately, Chennault blunted the effect of the Fourteenth 
Air Force's warnings by stressing, now the danger of a Japanese air attack, now 
the danger of a ground offensive, and by suggesting that the Chinese drive the 
Japanese from central China. The Japanese opened their own attack in mid- 
April, just after the Chinese, in response to the President's proddings, finally 
agreed to attack across the Salween into Burma. 


The Chinese Take the Offensive 

The Chinese decision of 12 April 1944 to attack across the Sal ween River 
came at a time when the Japanese had begun their drive on India, when Stilwell 
had decided to strike at Myitkyina, and when the Japanese had assembled along 
the Yellow River. As described by the Chief of Staff of the Chinese Army, 
General Ho, to General Hearn not later than 14 April, the Chinese plans had 
for their first objective the city of Teng-chung; for the second, Lung-ling. Their 
later actions and troop movements, the Chinese added, would depend on how 
the situation developed. 1 General Marshall, on receiving Hearn's message to 
this effect, told General Ho that the Chinese move might well be the decisive 
blow in the campaign to regain control of north Burma. 2 

The Chinese (and it must be remembered that this campaign would be 
fought by Chinese armies in China Theater under Chinese leadership) there- 
fore aimed at seizing two key points on the trace of the projected line of 
communications to China. Teng-chung, an old jade marketing center, lies 
almost due east and 124 miles from Myitkyina, on a rough but usable trail. 
Lung-ling is on the old Burma Road. Its capture would split the Japanese 

positions along the Sal ween. (Map 19*) 

The specific objectives which Stilwell's officers, working with the Chinese 
commanders of Y-Force, had recommended on 29 March and which presum- 
ably reflected StilwelPs views of that date, had been to "secure and hold the 
general line: Mongmit-Lashio-Takaw-Monglen" while blocking any Japanese 
invasion of Yunnan from the direction of French Indochina. The result of 
success in such a move would be a grand converging attack of Stilwell's five 
Chinese divisions from India and the Generalissimo's Y-Force from Yunnan 
that would meet somewhere deep in north Burma south of the Myitkyina- 
Bhamo area. 3 The trace of the Ledo Road would then be free of Japanese. 

That the offensive was not ordered until mid- April robbed the decision of 
some of the significance it might have had if the Chinese had made the decision 

1 Rad CFB 16100, Ho to Hearn for Marshall, 14 Apr 44. Item 2214, Bk 6, JWS Personal File. 

2 Rad WAR 23478, Marshall to Ho, 15 Apr 44. Item 2225, Bk 6, JWS Personal File. 

5 Plan Rainbow, for the Salween operation, is contained in Incl 1 to Ltr 0, Col Richard M. 
Sandusky, Deputy CofS, Y-FOS to COs, Traveling Instructional Gps 1, 2, and 3, 29 Mar 44, 
sub: Rainbow Opns Plan. OCMH. 

*For an explanation of Chinese unit symbols, see|note 7(2)|below. 

T -'-••itticic 

MAP 19 



to cross the Salween in December 1943 or January 1944. At that time about five 
months of good weather would have lain ahead. In February, StilwelPs mission 
to Washington told OPD that Stilwell thought a bold attack might lead to the 
enemy's yielding all Burma north of Mandalay. In that event, with the road 
open to China, current plans called for adding 1,350 tons of equipment to that 
already possessed by each Y-Force division. These munitions, mostly trucks 
and 105-mm. howitzers, would have placed a most formidable group of divi- 
sions at the Generalissimo's disposal. 4 But the months had gone by and in 
mid- April the monsoon rains were but thirty to forty days away. 

In April 1944 the American leadership saw the Salween situation as an 
opportunity created by the current state of the 56th Division, its strength 
depleted by detachments and stretched over 100 air-line miles. A strong Chinese 
blow at the thin line might crumble it and the Japanese grip on north Burma 
would be broken. 

Battleground Above the Clouds: The Salween Front 

The Salween or Lu Kiang (the name Salween is of Burmese origin) has cut 
for itself a deep gorge through land 9,000 feet above sea level. The river is 
rarely more than 200 feet wide, but its waters race exceedingly swift, deep, and 
cold through the lower extension of the Himalayas to win for it the Chinese 
name of "Angry River." During the spring thaw and later in the monsoon 
season the Salween swells into a torrent almost impossible to cross. During 
their hasty retreat in May 1942 the Chinese had destroyed the two bridges then 
crossing the Salween. 

If the Japanese 56th Division was to be driven from its line along the Salween 
and across the Burma Road, the Chinese would have to be ferried across the 
Salween at several points. Then the Chinese would have to cross the grain of 
the Kaoli-kung Mountains, their advance funneled into the mountain passes. 
Once through the mountains, they would be moving toward their objective, 
the Myitkyina-Bhamo-Lashio area. Apart from the Burma Road itself, four 
usable passes through the Kaoli-kung Mountains offer as many gateways to 
Chinese invaders of Burma. From north to south, these are: Hpimaw Hkyet at 
9,000 feet, the northernmost extension of the 56th Division's front; Ma-mien 
Kuan at 10,000 feet, thirty air-line miles south of the Hpimaw pass; Ta-tang-tzu 
and Hung-mu-shu, both at 10,000 feet, are traversed by pack trails between 
Pao-shan and Teng-chung. 

Fifteen air miles west of the Salween and running parallel to it through a 
valley of the Kaoli-kung Mountains is a similar but smaller stream, the Shweli, 
whose mountain waters run 6,000 feet above sea level. Between Kaochiao, a 

4 Min, 4th Conf, USAF CBI Offs with OPD WDGS, Washington, 12 Feb 44. OCMH. 



village at the western end of Ma-mien pass, and Chiang-chu, a small settlement 
at the terminus of Ta-tang-tzu pass, the Shweli cuts through a narrow valley. 
A secondary road, used for Japanese motor traffic in dry weather, links Kaochiao 
and Chiang-chu. This road continues to the southwest to become a natural 
avenue of approach to Teng-chung, the last large town on the China side of 
the border. West of the Shweli valley the terrain is not so high as that between 
the Shweli and Sal ween Rivers, but it is just as hard for any army to cross. A 
good trail, which offered the possibility of being made into a military road, ran 
between Teng-chung and Myitkyina, 124 miles north and west of Teng-chung. 

While the general ruggedness of the Kaoli-kung area is unbelievable, and 
the monsoon rains in their proper season greatly increased the difficulties and 
miseries of the campaign, it was the veterans of the 56th Division with their 
skillful deployment over the mountains and along the valleys who were the 
principal obstacle to the Y-Force in meeting Stilwell near Myitkyina. Elements 
of three regiments, the 113th, 146th, and 148th of the 56th Division, plus two 
companies of the 11 4th Regiment, 18th Division, at Hpimaw pass watched the 
long front. The commander, Lt. Gen. Sukezo Matsuyama, and 56th Division 
headquarters were at Mang-shih on the Burma Road. 5 

Twelve miles northwest of Mang-shih on the Burma Road was Lung-ling, 
strongly garrisoned since it controlled trails that branched north and south of 
the Burma Road. Some forty miles north of Lung-ling, the 148th Regiment held 
Teng-chung and turned that old town with its thick walls into a formidable 
bastion of the Japanese line. From Teng-chung, patrols moved up and down the 
Shweli valley, watched the passes through the Kaoli-kung, and frequented ferry 
sites along the Salween. At the center of the Japanese line the 113th Regiment 
thoroughly fortified Sung Shan, a multipeaked mountain, which dominated the 
site of the destroyed Hui-tung Bridge over the Salween and the first twenty- 
seven miles of the Burma Road west of the Salween River. South of Sung Shan, 
detachments garrisoned Ping-ka and Hsiang-ta, cholera-plagued villages which 
controlled trails reaching the first great bend of the Salween before it cuts the 
Burma border. At Kunlong Ferry, sixty-five air miles away, where two years 
before the Allies had worked on the Yunnan-Burma Railway, the 56th Division 
placed three companies of its reconnaissance regiment to hold the southern end 
of its line. 

The 56th Division depended on the Burma Road and the trails branching 
from it to link its forward elements with their principal supply base and head- 
quarters at Lashio. The Japanese trucks were vulnerable to air attack, but 
throughout 1943 and the first half of 1944 the Fourteenth Air Force was unable 
to stop Japanese traffic along the Burma Road. 6 

5 Although 56th Division was originally a part of 15th Army, it was placed directly under 
the control of Burma Area Army on 15 February 1944. Japanese Study 93. 

6 Japanese Study 93. 



Chart 6 — Schematic Order of Battle of the Chinese 
Expeditionary Force 

...i TiunibtFtd dni^ol'Cnsaf 1K1 ihit* 't^iminii fn d Chmew 
Thtpiodvd i|-trii"umbiro[lh*lhiiidF«9imtnl. THui.thtf 
'Attached 54lh Army, 
*Anocb«d let Amy. 
•Am^nnf Hq CEF.Moy 44. 
d A»r 9 n«J He. CEF. Ag 3 44. 

fern. (1) AG (Y-FOS) 111. KCRC. (5} S.. S., Mi„ >D n >„ Tdnl. 6 ~d Cher, J 

; n4 Iht difiiicn't dti>9<i<iHan by ihrtt. 
■ 57th. 

The Chinese Plan for the Salween Campaign 

The Chinese Expeditionary Force consisted of two group armies, the XI 
and XX. (Chart 6) The XI Group Army, commanded by Maj. Gen. Sung 
Hsi-lien, commanded the 2d, 6th, and 71st Armies. Headquarters, 6th Army, 
remained in the rear while the army's two organic divisions were attached to 
different group armies, the 2d Reserve Division to the 54th Army, and the 39th 
Division to 2d Army/ The XX Group Army, under Maj. Gen. Huo Kwei- 
chang, included the 53d and 54th Armies. Thus, at the start of the Salween 
campaign, Gen. Wei Li-huang commanded twelve divisions. Unfortunately, the 
Chinese Ministry of War had not brought these units up to strength, so that 
Wei's actual strength of 72,000 men was 40 percent below what it should have 

By the end of February 1944 the five Y- Force armies had each been issued 
540 .45-caliber submachine guns, 54 .55-caliber Boys antitank rifles, 162 60-mm. 
mortars, 72 rocket launchers, 12 75-mm. pack howitzers, and 54 Bren guns. The 

7 (1) Of the five Chinese armies, the 2d was transferred from west Kweichow Province and 
the 53d from Hunan. The rest had been stationed in Yunnan. See General Plan of Anakim, 19 
May 1943, cited note 9(2). (2) A Chinese army may be considered, on the basis of its authorized 
strength, to be the equivalent of an American army corps and is, therefore, indicated on the 
maps by a corps symbol. The actual numerical strength of a Chinese army, however, was closer 
to that of a U.S. division, which it resembled in organizational concept. Its divisions were organic 
to it, just as three infantry regiments are an organic part of the U.S. infantry division. 

ammunition for these weapons was ample to start the offensive and continued 
supply presented no problem to Y-FOS supply members. 8 

Estimating that the 36th Division was firmly entrenched along the Burma 
Road, and that it could easily bring reinforcements up the road to the Japanese 
forward areas, the Chinese decided to strike first at the flanks of the long Japa- 
nese line. When their task forces had successfully made their crossings of the 
Salween, they would move inland, then close in on the center of the Japanese 

8 (1) Min cited fn^l ( 2 ) As of 20 April 1944 lend-lease weapons and ammunition issued 
to the Y-Force by Y-FOS G-4 included: 

Weapons Ammunition 

75-mm. pack howitzers 244 7 5 -mm. high-explosive 378,334 

37-mm. antitank guns 189 75-mm. smoke 4,910 

Boys antitank rifles (.5 5 -caliber) 536 37-mm. armor-piercing 55,020 

60-mm. mortars 1,238 37-mm. high-explosive 11,260 

Rocket launchers Ml Al 395 . 5 5-caliber armor piercing 623,980 

Thompson submachine guns (.45-caliber) 5,631 60-mm. high-explosive mortar 630,037 

Bren machine guns (.303-caliber) 603 2.36-inch antitank 2,880 

.45-caliber ball 18,640,400 

. 303-caliber ball 25,232,000 

7.92-mm. ball 164,551,500 

In addition Y-FOS furnished 475 jeeps, 1,999 radio sets, and 420 pounds of napalm gasoline 
thickener. Ltr AG (USAF CBI) 400.3591 and Incls, Gen Evans, Deputy CofS, to Stilwell, 15 
May 44. Stilwell Documents, Hoover Library. 



position from north and south. When the envelopment was judged almost 
complete, then the Chinese proposed to attack down the Burma Road. It was 
expected the Japanese would then be forced to retire from Yunnan. In falling 
back south and west, they would probably yield the town of Teng-chung, 
opening trails to Myitkyina and Bhamo. 9 

The XX Group Army was assigned the sector north of the Burma Road; 
the XI Group Army, the Burma Road and the Japanese positions to the south. 
General Wei's plan called for the XX Group Army to send three reinforced 
regimental combat teams across the Salween at as many points. When each had 
made good its foothold, reinforcements would follow and that portion of the 
attack would be under way. 

The northernmost pass, the Hpimaw pass, was left by the Chinese to the 
irregulars who had harassed the Japanese there since February; no effort 
through it was scheduled. Twenty-five miles to the south was Ma-mien pass. 
The Chinese ordered the 54th Army to send the 593d Regiment, 198th Divi- 
sion, across the Salween on D Day. The 593d Regiment was to filter through 
the Ma-mien pass, using secondary trails to avoid Japanese patrols, and enter 
the Shweli valley which, it will be recalled, was a north-south corridor behind 
the main Japanese defenses. Once the 593d was in the Shweli valley, the 
Chinese believed they would be safe from Japanese counterattacks in the 
Ma-mien pass. Following the 593d Infantry would come the remainder of the 
198th Division to clear the Japanese from the Ma-mien pass area and to drive 
on into the Shweli valley. The 593d Infantry's crossing of the Salween would 
be aided by the 2d Reserve Division which was to move south and seize the 
village of Hai-po halfway down Ma-mien pass to the Salween's banks, and 
thus prevent the Japanese from bringing up reinforcements to oppose the 
593d's crossing. 

Seventeen miles south of lofty Ma-mien pass, the main force of XX Group 
Army was to clear the Ta-tang-tzu pass, the third gateway through the Kaoli- 
kung Mountains. This pass was held by the 148th Infantry (less the 1st and 2d 
Battalions). At the western end of the pass was the principal Japanese strong- 
hold in the immediate area, the fortified village of Chiang-chu. On D Day, the 
36th Division of the 54th Army, plus one regiment of the 116th Division of 
the 53d Army, was to make the Salween crossing. With the bridgehead estab- 
lished, the rest of the 53d Army would follow. Three Chinese divisions would 
force their way through Ta-tang-tzu pass. This would place them in the Shweli 
valley, ready to link with the 198th Division, and to drive south down the 
Shweli valley as part of a great pincers on the Japanese. 

9 (1) Stilwell's Mission to China [ Chart 8.| (2) Y-Force, with the advice of Y-FOS, had begun 
planning for the Salween operation eariy in iy43. The product of its staff work was Plan Rainbow. 
which the Generalissimo never approved. The General Plan of Anakim (Y-Force Project), 19 
May 1943, OCMH, contains the general scheme for re-equipping the Y-Force divisions from 
Chinese and U.S. contributions and the initial Plan for the conduct and objectives of the Salween 
operation. See Stilwell's Mission to Cbina \Ch. VIII. | ( 3) Plan Rainbow, Dec 43 and Feb 44 
versions, AG (Y-FOS) 381, KCRC. 



The shortest route from Pao-shan, the Chinese headquarters and communi- 
cations center, to Teng-chung is that which crosses the Salween at the site of 
the Huei-jen Bridge, but the pass on the Burmese side was so difficult that the 
Chinese intended to confine themselves to preventing its use by the enemy, 
rather than attempt to take it as part of their own attack. A regimental combat 
team reinforced, from the New 39th Division, was ordered to cross the Salween 
on D Day at three ferry sites three to five miles above the Huei-jen Bridge, 
which was firmly held by a Japanese battalion, then swing south on the bridge 
site. The rest of the New 39th Division would follow the combat team across. 

Next in order of the gateways to Burma is that through which the Burma 
Road passes, but since the Japanese were believed to hold it in force, the 
Chinese plan called for postponing attack on it until the double envelopment 
was well under way. Therefore the next Chinese assault was to cross the Salween 
eleven miles south of the village of Ping-ka. Elements of the 71st and 2d Armies 
would follow and converge on Ping-ka, whose capture was expected to open 
the way up the Ping-ka valley into Burma. While Ping-ka was under attack, 
other elements of the 71st and 2d Armies would bypass it and swing northwest- 
ward toward the Burma Road and the important towns of Mang-shih and 
Lung-ling. Their capture would clamp a Chinese roadblock firmly on the Burma 
Road, cutting off some of the strongest Japanese positions. Once these two 
towns were under attack, the rest of the XI Group Army would cross the 

The success of the operations outlined above would precede General Wei's 
commitment of his reserve to an attack directly down the Burma Road. 10 

Such then was the Chinese Expeditionary Force's plan for an offensive across 
the Salween, a double envelopment that would scoop out the Japanese from 
their fortified positions. The plan was good, but the hour was late and this 
would be a great handicap. As far back as February 1943, the Chinese had 
agreed to launch their attack in October 1943, thus insuring six months of good 
weather. Over this same period, Stilwell and his subordinates in China had 
worked closely with Y-Force to prepare it for its projected role in the North 
Burma Campaign. 11 But the Chinese had let the months slip by and now in 
May 1944, on the eve of the monsoon rains, Y-Force would demonstrate the 
extent to which it had profited from American advice and aid. 

The American Contribution 

The American contribution to China's offensive took two forms, advice and 
logistical support. The Americans had no command functions in Y-Force, but 
since April 1943 they had been advising in training, in supply, and in opera- 

10 (1) Japanese Study 93. (2) CM-IN 10243, Ho to Marshall, 14 Apr 44. (3) Rad 313, 
Dorn to Hearn, 19 Apr 44. Item 2244, Bk 6, JWS Personal File. (4) Ltr, Dorn to Stilwell, 19 
Apr 44; Memo, Dorn for StilwelL 9 Mav 44. SNF 35. 

11 Stilwell 's Mission to Cbina,\U^. VIII. | 



AN AMERICAN LIAISON OFFICER studies Japanese positions across the Salween 
River before the crossing. 

tions. Studying the terrain, the Japanese positions, and the Chinese capabilities, 
the American liaison and instructional groups since early 1943 had been urging 
the Chinese commanders to accept certain basic principles. The first of these 
reflected the fundamental weakness of the Japanese position, that no matter 
how devoted the individual Japanese soldier or how skillfully he had fortified 
his position, the 56th Division had to guard over 100 air-line miles. Therefore 
the Americans had been urging the Chinese to prepare to infiltrate through the 
inevitable gaps in the Japanese line. As the situation suggested, all Chinese 
units should be prepared to break up into small, highly mobile, self-sustaining 
combat teams. Each combat team should be prepared to operate independently 
for several days. Each should be indoctrinated with the principle of moving 
ahead regardless of Japanese attempts at envelopment. If the Japanese suc- 
ceeded in placing themselves behind a Chinese unit, that unit should in turn 
attack the Japanese rear and try to outlast them. In their advance, the Chinese 
should keep themselves deployed in depth so that there would always be com- 
bat units in the rear able to deal with infiltrating Japanese elements. 

Warfare on the mountain trails would, the American advisers stressed, limit 
the Chinese Expeditionary Force to the use of infantry supporting weapons— 



mortars, machine guns, and flame throwers— augmented by a few pieces of pack 
artillery. Weapons and ammunition of this nature could be airdropped to keep 
the advance under way. Few occasions would arise in which artillery would be 
used as a battalion to support a Chinese division. This belief led to preaching 
the use of individual batteries working with mortars and automatic weapons 
as integral parts of small combat teams. 

Pack animals and coolie labor, the Americans foresaw, would provide the 
greater part of the supply transport in the early phases of the campaign. Carts 
could assist, but motor vehicles would be confined to a few routes. The Chinese 
were told that, because of the terrain and the probable possession of air superi- 
ority, troops must learn to co-operate with friendly aircraft attacking Japanese 
positions and in receiving airdrops. The monsoon rains would of course hinder 
air supply so the Chinese were urged to capture Japanese landing strips and 
sites adaptable to airdrops. 12 

After General Ho formally undertook to cross the Salween, the American 
operational staff with the Chinese divisions, or Y-FOS, had three weeks within 
which to complete its preparations. In making these final arrangements, Dorn 
stressed again and again the four responsibilities he had accepted. To meet them 
plus his other missions he devised an organization which closely paralleled that 
of the Chinese Expeditionary Force. On 21 April, General Wei outlined the 
liaison mission that Y-FOS would have in relation to his Chinese troops. One 
point was clear— no American had command over the Chinese. Y-FOS would, 
however, continue to (1) assist in training, (2) assist in supply, (3) perform its 
own administration, (4) exchange intelligence with the Chinese, (5) furnish 
air-ground liaison, (6) report to the Chinese Expeditionary Force "the needs of 
the Chinese front line troops so that CEF Headquarters may decide something 
to assist them." 13 

On 29 April, Dorn established the Field Headquarters, Y-FOS, with a G-2, 
G-3, G-4, and chief of staff. This group was to accompany General Wei's 
headquarters. Meanwhile, Y-FOS teams, plus attached medical units, either 
joined Chinese units or augmented American liaison groups already in the field. 
Because most of his divisions already had U.S. traveling instructional groups 
with them, Wei allowed Y-FOS to expand the teams in order to reach the regi- 
mental level. Each Y-FOS team varied in size from six to twenty Americans 
and usually included infantry, artillery, engineer, ordnance, signal, quartermas- 
ter, and veterinary personnel. Some 100 signal communication enlisted person- 
nel were among the Y-FOS teams. Portable surgical and field hospitals and 
veterinary detachments were the major portion of U.S. table of organization 

12 (1) Hist Rpt, Y-FOS (1 Jan-24 Oct M\ r Hereaf ter Y-FOS 1944 Hist Rpt.) AG 
(Y-FOS), KCRC. (2) Stilwell's Mission to China lC.h. VIIT.i m Memo, Dorn for Stilwell, 2 Feb 
44. AG (Y-FOS^ 381. KCRC. (4) Plan cited n. 9(2). 

15 (1) See Ch. VIII above. (2) Quotation from Memo, Tsen 2114, Wei for Dorn, 21 Apr 44. 
AG (Y-FOS) 319.1, KCRC. 



units with the Chinese Expeditionary Force. U.S. antiaircraft batteries were 
stationed at vital bridges and airfields. 14 

In rear areas, G-4 of Y-FOS and SOS expedited delivery of food, ammuni- 
tion, and equipment. These were received at the Kunming and Yun-nan-i 
depots, and brought forward from them by trucks along the Chinese portion of 
the Burma Road, by pack and coolie train over the mountains, and later by 
airdrop on the battlefield. Because there were no accurate maps of the Salween 
area, the Fourteenth and Tenth Air Forces flew photo-reconnaissance missions 
to produce aerial mosaics. American engineer troops, always convinced that the 
D-Day crossings could be successfully made, planned to ferry the five assault- 
ing task forces over the river and gave Chinese engineers intensive training in 
the use of rubber boats. Daily practice was conducted across the turbulent 
Mekong River. 15 

Reconstruction and improvement of the Burma Road from Kunming to 
Pao-shan was a major preoccupation, and Y-FOS engineers were detailed to 
co-operate with a Chinese agency, the Yunnan-Burma Highway Engineering 
Administration, in planning the work. On 1 May SOS assumed this responsi- 
bility from Y-FOS, and on 15 June organized the Burma Road Engineers. 
Enjoying its own table of organization, this unit was attached to SOS. General 
Dorn gave strong support to every phase of the Sino-American effort to rebuild 
the vital highway from Pao-shan to the Salween, including plans for recon- 
structing the destroyed Hui-tung Bridge which had carried the road over the 
Salween 's dramatic gorge. Dorn gave the Burma Road Engineers first priority 
on Y-FOS personnel for its engineering staff. 16 

In working out the details of tactical air support and air supply with the 
Fourteenth Air Force, Y-FOS received complete co-operation. The 69th Com- 
posite Wing had the responsibility of supporting the China Expeditionary 
Force's ground effort. Sergeant pilots of the 19th Liaison Squadron were 
attached to Y-FOS for courier service in their little L-5's. Since he had felt 
obliged to assure the Chinese that the United States would contribute the factor 
of air supply to their offensive, Dorn asked Stilwell for a C-47 squadron plus 
necessary personnel. Though the request could not be filled at once, the 27th 
Troop Carrier Squadron joined the Fourteenth Air Force in late May. 17 

Beginning the Offensive 

Following the preparations from afar, the Generalissimo telephoned Gen- 
eral Wei on 27 April to make a few last-minute changes in the Chinese order 
of battle and to set D Day. The Generalissimo also demanded that his com- 

14 (1) GO 3, Hq Y-FOS, USAF CBI. (2) Y-FOS 1944 Hist Rpt. 

15 Y-FOS 1944 Hist Rpt. 

16 ( 1) SOS in CBI, App. 1, SOS in China, Burma Road Engrs. (2) Interv with Col Robert F. 
Seedlock, former CO, Burma Road Engrs, Dec 1950. 

17 Y-FOS 1944 Hist Rpt. 



manders "succeed— or else!" On 9 May Dorn notified Stilwell that the Salween 
crossings were to be made on the night of 10 or 11 May, that hopes were high 
of reaching Myitkyina before Stilwell did. 18 

The crossing of the Salween River, an operation the Chinese had regarded 
with justified apprehension, since Japanese machine guns and artillery might 
turn the Salween line into a most formidable obstacle, went with clocklike pre- 
cision. The Japanese had decided not to defend the crossing sites, but to place 
their main line of resistance along the main ridge line some ten miles west. 
From decoded Chinese radio messages and from the forward movement of Wei's 
headquarters, the Japanese had concluded the offensive was imminent, and 
made their last-minute preparations, but these did not include contesting the 
landings. 19 

Assembling ten miles east of the Salween, the 198th Division (Ma- mien 
pass sector) received an army order on 9 May to move up to the crossing site. 
During the night of 11 May, Chinese engineer companies, supervised by seven- 
teen Y-FOS soldiers, commenced the ferrying operation. Throughout the 
moonlit night and on until noon of the following day, engineers shuttled the 
rubber boats, bamboo and oil-drum rafts, and similar expedients across the 
Salween's swirling eddies and currents. So strong were the Salween's currents 
that it took four engineers to paddle but four infantrymen and their impedi- 
menta across the river at a time. Larger ferry boats carried pack animals and 
artillery. During late afternoon of 12 May this force of Chinese attacked its first 
objective. 20 

On 11-12 May, the three regiments of the 36th Division plus the 346th 
Regiment of the 116th Division (Ta-tang-tzu pass sector) successfully crossed 
the Salween at Meng-ka ferry. From D Day on, the ferrying operations went 
smoothly, "elements behaved quietly and obeyed instructions . . . through- 
out." 21 

The regimental combat team assigned to hold the Huei-jen Bridge area, 
which comprised the 115th Infantry Regiment, plus a battalion of the 116th 
with artillery and service troops, was ferried across the Salween without inci- 
dent on the night of 10-11 May. The Chinese did not receive heavy fire until 
morning. 22 

The southernmost crossings, those near Ping-ka, were rather elaborate in 
conception. Before D Day, four companies of the 9th Division, 2d Army, 
slipped across the Salween and moved close to the village of Ping-ka to keep 
Japanese patrols from leaving it to go to the river's edge. On D Day, the 

18 (1) Rad CYF 407, Dorn to Stilwell, 27 Apr 44. Item 2273. Bk 6. TWS Personal File. (2) 
Memo, Dorn for Stilwell, 28 Apr 44. SNF 35. (3) Memo citec jn. 1(1(371 

19 Japanese Officers' Comments, p. 27; App. 1, Col Fujiwara. 

20 Y-FOS 1944 Hist Rpt, App. V, Rpt, U.S. Detachment, 54th Army Engr Co, 18 Aug 44. 

21 (1) Ltr, Col John H. Stodter, CO, U.S. Ln Gp, 53d Army, to CG, Y-FOS, 15 Oct 44, sub: 
Opns of 53d Army. (Hereafter, Stodter Report.) OCMH. (2) Interv with Col Stodter, 11 Apr 47. 

22 Rpt, Maj Lawrence W. Beilson to Dorn, 10 Aug 44, sub: Rpt on Opns of 39th Div. AG 
(Y-FOS) 317.1, KCRC. 



Chinese made two crossings, one in the area eleven miles below Ping-ka, the 
other, seven miles northeast. Below Ping-ka, the Chinese sent across the 228th 
Infantry as a regimental combat team reinforced, from the 76th Division, 2d 
Army. Above Ping-ka, the attack was begun by a similar combat team built 
around the 264th Infantry, 88th Division, 71st Army. The two task forces were 
to converge on Ping-ka from north and south. Here as everywhere else fighting 
did not begin until the Chinese were well over the river. 23 

Pushing Through Ma-mien Pass 

Fighting in Ma-mien pass, where the Chinese were trying to clamp the 
northernmost part of their pincers in place, began on the afternoon of 12 May 
when a battalion on the right flank of the 198th Regiment attacked the first 
Japanese outpost. Making good progress, by dark the battalion had occupied 
several Japanese pillboxes and part of the trench system. All that night, the 
Japanese quietly filtered down from a nearby ridge and assembled near the 
Chinese position. Attacking at dawn, they surprised the Chinese and almost 
wiped them out before aid came. The Chinese battalion commander was killed, 
also 2d Lt. Kirk C. Schaible, the first Y-FOS liaison officer to die in action. 
Seeking to cover an exposed flank and find a better field of fire, Lieutenant 
Schaible left cover and was instantly killed by a Japanese rifleman. Chinese 
reinforcements restored the situation, and Japanese resistance in the immediate 
area ceased on the next day, 13 May. 24 

The Japanese defenders at Ma-mien pass had been the 2d Battalion, 148th 
Regiment, under a Colonel Kurashige. Kurashige had moved his men into the 
pass in January. He had relied on patrols to watch the ferry sites, while small 
garrisons in strongpoints held the mountain trails. 25 

The Chinese established their line of communications across the Salween 
on 29 May, when a footbridge was repaired to permit supplies and pack animals 
to cross the river. Telephone wire had been strung across the river two days 
earlier. The Chinese were then no longer dependent on their first means of 
communication with the rear, L-5 liaison aircraft operating from a landing strip 
which Chinese engineers had hacked from the mountainside between 12 and 

23 (1) Memo citec j n. 10(4).| (2) U.S. liaison reports on 2d Army's role in the Salween cam- 
paign consist of the following:, Col Robert C. Lutz, CO, U.S. Ln Gp, 2d Army, to CG, Chinese 
Combat Comd, USFCT, 5 Feb 45, sub: Opns of 2d Army During Salween Campaign; Rpt, Capt 
Eugene D. Hill, 2d Army Ln Gp, 26 Aug 45, sub: Brief History of 2d Chinese Army; Rpt, Maj 
Thomas G. Maris, Jr., CO, U.S. 76th Div Ln Team, 16 Dec 44, sub: Opns of 76th Div; Rpt, Maj 
William L. Lowry, CO, U.S. 9th Div Ln Team, 1945, sub: Opns of 9th Div. AG (CCC) 314.7, 

24 (1) GO 69, Hq Chinese Combat Comd USFCT, 7 Nov 45. (2) The best U.S. Army source 
for a detailed combat story of the Salween offensive is a daily G-2, G-3 Journal kept by Head- 
quarters, Forward Echelon Y-FOS, at Pao-shan between 11 May 1944 and 28 February 1945. The 
journal consists of two volumes. A carbon copy is filed with Zebra Force Records, KCRC. 
(Hereafter Y-FOS Journal.) 

25 Japanese Study 93. 


20 May- Unfortunately the footbridge proved inadequate to sustain the Chinese 
in Ma-mien pass, and air supply was required, 

Having beaten off the first Japanese counterattack, the 592d and 594th Regi- 
ments of the 198th Division began clearing out Japanese strongpomts while 
the 593d moved west over mountain byways to emerge into the Shweli valley 
on 16 May, near the village of Kaochiao. Promptly, the Chinese attacked the 
Japanese garrison of Chiao-tou-chieh which, surprised and momentarily pan- 
icked, took to its heels. But the Japanese panic was of brief duration, and 
stouter resistance kept the 593d from moving farther south and down the 
Shweli valley. The 593d Regiment halted, posted guards to close off the western 
end of Ma-mien pass, arid waited for the main strength of the 54th Army to 

Within Ma-mien pass itself, the survivors of the 2/148 fell back on the 
fortified village of Chai-kung-tang. Already present were the 2d and id Battal- 
ions of the I 13th infantry, rushed north from Ping-ka and Bharno as soon as the 
J 6th Division had made its estimate of the Chinese intentions. Chai-kung-tang, 
while the Japanese held it, would close the narrow Ma-mien pass to Chinese 

Well fortified, the Japanese resisted staunchly, and only desperate fighting 



by the Chinese cleared them from the pass by 13 June. When the Chinese could 
not take a bunker by their usual technique of frontal attack, they would request 
the American air liaison officer to call up support from the 51st Fighter Group 
of the AAF at Yun-nan-i. Y-FOS personnel taught the Chinese how to prepare 
ground indicators to guide the P-40's on their target runs. Using rockets and 
fragmentation and demolition bombs, Chennault's pilots acted as artillery for 
the Chinese and flew close support missions as often as weather permitted. 

The weather grew progressively worse as the monsoon rains closed in. In 
those high altitudes the rains became blinding sleet and fog, an added misery 
of war for the poorly clad Chinese soldiers. The rains threatened to wash away 
the Chinese line of communications, and as the siege of Chai-kung-tang went 
on day after day, the coolie pack trains proved incapable of keeping the 54th 
Army's supplies from nearing the starvation level. Fortunately, the 27th Troop 
Carrier Squadron arrived at Yun-nan-i on 26 May and was in action two days 
later dropping tons of ammunition, rice, and much-needed raincoats. This was 
a welcome relief, but a brief one, for the weather from 3 to 10 June was so bad 
that flying was impossible. Chinese and Americans did what they could on what 
was left of their rations, supplementing them by bamboo shoots and unwary 
Burmese livestock. On one occasion, the American liaison team was able to 
supplement its rations by shooting a mountain tiger and converting it into 

With TNT charges dropped by the 27th Troop Carrier's C-47's when the 
weather cleared, the Chinese blew up the last pillboxes at Chai-kung-tang on 
13 June. When the last shots had been fired and the Chinese farmer boys of the 
54th Army reported the area secure, there was bewilderment at finding only 75 
Japanese bodies in defenses that must have been manned by at least 300 men, 
and shock and nausea when the Japanese kitchens revealed how the defenders 
had been able to prolong their stay. Pitiful and ghastly evidence showed that 
the Japanese had resorted to cannibalism when their rations failed. 

With their food stocks exhausted, the defenders had been ordered by 
Colonel Kurashige to escape at night and to fall back into the valley of the 
Shweli. Annihilation was inevitable if the Japanese garrison clung longer to its 
defenses, and Kurashige wanted its survivors to live and fight another day. 

With Chai-kung-tang firmly in Chinese hands, with clearing skies permit- 
ting the C-47's to return again with their cargoes of food and bullets, the 54th 
Army, victors of Ma-mien pass, began moving on into the Shweli valley. 26 

Clearing Ta-fang-fzu Pass, 11 May-12 June 1944 

The mission of the 53d Army was to drive through Ta-tang-tzu pass and 

unite with the 54th Army coming down from Ma-mien pass in the north. As 

26 (1) Y-FOS Journal. (2) Rpt, Capt Arthur M. Murphy, U.S. Ln Off, 592d Inf, 198th Div, 26 
Jul 44. AG (Y-FOS) 371.1, KCRC. (3) Hist Rpt, Hq 69th Composite Wing, Fourteenth AF, 29 
Jun 44. USAF Hist Div. (4) Memo, Col Frederick W. Hein, CO, U.S. Ln Gp, 54th Army, for CG, 
Y-FOS, 17 Oct 44, sub: Ln Off Rpts on Salween Campaign. OCMH. (5) Japanese Study 93. 



the 53d Army moved over the Salween, American liaison personnel with it 
were dismayed to learn that XX Group Army had ordered no Japanese position 
was to be bypassed. This meant that every Chinese unit north of the Burma 
Road would be forced to dig the stubborn Japanese from their mountain 
strongholds rather than move past them on into Burma. This order from XX 
Group Army had grave implications for the success of the Chinese effort. 

By dusk of 12 May the 36th Division had surrounded the Japanese outposts 
in the eastern end of the pass. When darkness came, the 36th went into 
bivouac. That night, the Japanese attacked vigorously, overrunning the division 
command post and causing the flustered 36th Division to fall back to the 
Salween. At dawn, the 53d Army commander, Maj. Gen. Chou Fu-cheng, 
pushed a regiment across the Salween and restored the situation by attacking 
the Japanese flank. General Chou was an aggressive and tenacious fighter, 
whom his Manchurian soldiers had nicknamed Old Board-Back, and who had 
the reputation of never having yielded an inch to the Japanese. But even Chou 
could not immediately restore the morale of the 36th Division, which for some 
weeks took no further part in the Ta-tang-tzu fighting, and the rest of the 53d 
Army had to bear the burden of clearing the pass. 

Though the smoothness with which the 53d Army brought its reinforce- 
ments across the Salween drew praise from American observers, its assaults in 
the eastern part of the pass fortifications were praiseworthy only for the wasted 
courage of its troops: 

In view of the enemy's defensive attitude and our superior strength, American liaison 
officers urged the use of a small continuing force and a strong encircling movement to cut 
the trail . . . behind the Japs. Nevertheless orders were received for direct attacks on the 
prepared positions . . . the 11 6th Div. to attack from the north and the 130th Div. from 
the south. Several days were wasted and heavy losses incurred ... in suicidal charges by a 
succession of squads against enemy pillboxes. Teamwork in use of weapons and supporting 
fires and the use of cover were conspicuously lacking . . . most casualties resulted from 
attempts to walk or rather climb up through inter-locking bands of machine gun fire. As a 
demonstration of sheer bravery the attacks were magnificent but sickeningly wasteful. Some 
platoon leaders were killed within one or two meters of the enemy embrasures and several 
of the best company and battalion commanders were killed and wounded in personal leader- 
ship of their troops. A general coordinated assault might have overrun the positions by sheer 
esprit and weight of numbers but adjoining or supporting units would idly watch some 
single squad or platoon get mowed down in a lone advance then try it on their own 
front. . . ? 1 

On days when the overcast disappeared the 51st Fighter Group struck at 
the Japanese bunkers, but the aerial artillery was not too effective against the 
Japanese bunkers of logs and concrete. Meanwhile, in order to support the 
valiant Chinese frontal assaults, Y-FOS personnel radioed American depots at 
Yun-nan-i for more ordnance materiel, for Chinese crews sometimes wrecked 
their weapons through misuse or lack of maintenance. Because of Chinese 

27 (1) Stodter Rpt. (2) Interv citec ln. 21(2*1 



AIR-SUPPLY DROPS like the one shown above were the key to success in Burma 

advances some miles to the north, the Japanese decided to thin out their garri- 
son and constrict their holdings in the Ta-tang-tzu pass area. Their withdrawal 
on the night of 22-23 May permitted an eight-mile advance by the Chinese. 

Over 28-31 May the 27th Troop Carrier Squadron replenished the 53d 
Army's stocks. Deliveries were made at so fast a rate that the airdropping crews 
were sometimes obliged to assist the packers, who found themselves unable to 
keep pace. 

To clear the remaining Japanese from Ta-tang-tzu pass, the 116th and 130th 
Divisions continued the attack. The 36th Division, which had been so roughly 
handled in the opening days of the campaign, was ordered to bypass the Japa- 
nese in the Ta-tang-tzu area, and to make its way sixteen miles northwest, to 
where the 54th Army, after clearing Ma-mien pass, was stalemated in the upper 
Shweli valley. The 36th Division found its way open because early in June 
Colonel Kurashige had been ordered by the 36tb Division to send the 3/ 148th 
to the Japanese northern flank. 

Behind the front, Chinese and Americans worked at communications and 
supply. Chinese engineers laid lend-lease telephone wire to open communica- 
tions with General Wei's Pao-shan headquarters. At the pace of the coolie's 



slow climb, supplies and pack artillery began to filter through to the upper 
Shweli. The efforts of the coolies, strenuous though they were, could not yield 
results comparable to the thirteen C-47's, whose crews began to feel they knew 
the Kaoli-kung peaks and valleys as well as they knew their hometown corner 
drugstores. Numbered airdropping sites were now well past the twenty mark. 

In addition to the obvious contribution of air supply, the Americans found 
themselves making two other important contributions to the campaign. Liaison 
radio teams forwarded considerable intelligence, air target, and supply data to 
Pao-shan, helping Wei and Dorn co-ordinate the Sino- American effort. Y-FOS 
liaison officers also reported tactical and logistical mistakes the Chinese made 
so that Wei and Dorn could take remedial action. Gradually, Sino-American 
co-operation in the field began to grow effective. 

But at this early stage of the campaign, reports from the American liaison 
teams were not always cheerful. Americans observing the Ta-tang-tzu and 
Ma-mien actions found that Japanese fire was accurate and economical, and 
that the enemy's use of camouflage and concealment approached perfection. 
The Japanese revealed no disposition to surrender though they were heavily 
outnumbered, often surrounded, and had neither air support nor air supply. 
On the other hand, the Americans reported that the Chinese endlessly wasted 
manpower and ammunition in costly frontal attacks. They reported that rela- 
tions with the Chinese were not always as friendly as had been hoped, and 
they believed there would have to be better co-operation between Chinese and 
Americans if the Japanese were to be defeated. The Chinese were described as 
merely tolerating the Americans' presence and as paying little attention to their 
advice. The liaison personnel freely admitted their own shortcomings, and by 
their reports suggested that patience was the most important quality for a 
liaison officer to cultivate when dealing with the Chinese. 

Most Americans liked the aggressive spirit and tenacity of General Chou, 
the 53d Army's commander of picturesque nickname. To Chinese officers like 
Chou, the offensive was theirs to plan and to fight, and the Americans were 
merely guests to be shielded from harm. While liaison officers appreciated the 
kindness of the Chinese in giving them the status of guests with all this im- 
plied of concern for their safety and comfort, they were appalled to observe the 
degree to which these same Chinese commanders considered their soldiers ex- 
pendable. Liaison officers were horrified to learn that a company commander 
could execute a soldier but that it took a group army order to shoot a horse or 
mule. 28 A month's experience with the way the Chinese proposed to fight the 
Salween campaign provided a clue to the difficulties the Chinese and Amer- 
icans would have to face and surmount before General Wei's soldiers could 
meet with General Stilwell's. 

28 (1) Hist Rpt, 27th Tr Carrier Sq, Fourteenth AF, May-Oct 44. USAF Hist Div. (2) Stodter 
Report. (3) Y-FOS Journal. (4) Ltr with atchd rpts, Col Harry A. Buckley, CO, U.S. Ln Gp, XX 
Group Army, to CG, Y-FOS, 1 Nov 44. OCMH. (5) Japanese Officers' Comments, p. 29. 



Securing the Huei-jen Bridge Area, 11 May-14 June 1944 

The regimental combat team of the 115th Regiment reinforced by a bat- 
talion of the 116th Regiment, which was to take and hold the Huei-jen Bridge 
area for the Chinese, made its crossing at sites five to nine miles above the 
bridge. From the crossing site, the Chinese task force swung south to drive the 
1/113, commanded by its regimental commander, from the immediate area of 
the Huei-jen Bridge. Behind it came the rest of the New 39th Division. By 17 
May advance elements of the combat team were in the village of Hung-mu-shu, 
well behind the Japanese, and only twenty air-line miles from the key city of 

The Japanese soon recovered from the initial surprise of the Chinese of- 
fensive and the 1/113 drove the Chinese from Hung-mu-shu. The Japanese 
exploited their success and pushed the entire New 39th Division back against 
the Salween. The Chinese collected themselves, renewed the attack, and by 27 
May had one element back within five miles of Hung-mu-shu, while the 115 th 
Infantry took hill positions overlooking the enemy's defenses at the bridge 
site and engaged the Japanese artillery. The monsoon rains seemed to the Chi- 
nese and American liaison officers to be confining both sides to their positions. 
In reality, the Japanese, in line with their tactics of a mobile defense, withdrew 
most of the 1/113 and sent it north to attack the Chinese 53d Army near 
Ta-tang-tzu pass, leaving only a few men behind. 

The pressure exerted by the 53d Chinese Army in the north in effect ended 
the Japanese defense of the Huei-jen Bridge area, for the Japanese were not able 
to return the 1/113. On 12 June the Chinese 115th Infantry was back in Hung- 
mu-shu, after killing thirty of the Japanese rear guard. Pushing west, Chinese 
patrols occupied a pass on the Hung-mu-shu-Teng-chung trail, opening a route 
to the Shweli valley. 

Instead of being directed on Teng-chung, the New 39th Division was sent 
south on a trail paralleling the Salween. Its orders were to unite with elements 
of the New 28th Division, 71st Army, that were attacking the Japanese strong- 
hold of Sung Shan, which controlled the Hui-tung Bridge area. 29 

Driving the Japanese Rear Guards From the Shweli Valley 

The Japanese withdrawal of several units from the northern flank in the 
upper Shweli valley to bolster the southern flank near Lung-ling left Colonel 
Kurashige's Japanese less able to hold the 53d and 54th Armies in the north. 
As the Japanese positions at Chieh-tou, Chiao-tou-chieh, Wa-tien, and Chiang- 
chu, along the upper Shweli, began to fall, prospects of a more speedy capture 
of Teng-chung by the Chinese began to appear. On 1 June units of the 54th 

29 (1) Rpt citec | n. 22.\ {2) Y-FOS Journal. (3) Japanese Study 93. (4) Japanese Officers' 
Comments, p. 31. 



Army, from the 2d Reserve and 36th Divisions, emerged in the Shweli valley 
from Ta-tang-tzu pass to join the 593d Regiment from the Ma-mien pass. Next 
day the Chinese occupied Chieh-tou village and invested Chiao-tou-chieh with 
two regiments of the 36th. Once 53d Army patrols commenced operations in 
the Shweli valley, they took the advice of Y-FOS officers and donned Burmese 
clothes. So disguised, they found it easy to enter Japanese-held villages. When 
taken by surprise, the Japanese were willing to abandon many of their outer 
defenses. On 14 June the Japanese quit Chiao-tou-chieh, leaving many stores 
to the 2d Reserve and 36th Divisions. 

The garrison of Chiao-tou-chieh retreated in two columns, each pursued 
by a Chinese regiment. One Japanese column crossed to the west of the Shweli 
valley and marched westward to reach a trail that would take it to Teng-chung. 
The other Japanese column, its strength raised to perhaps 500 by patrols and 
outposts it picked up as it went, withdrew southward on Wa-tien, twenty-five 
air-line miles above Teng-chung. The Chinese snapped at its heels until it 
entered the defenses of Wa-tien, then began to prepare another siege. 

At this point the 36th Division ordered the 2 /148th to fight a delaying action 
toward Teng-chung and to release the 3/l48th to meet the growing Chinese 
threat toward the 36 th Division's southern flank. The next Chinese objectives 
in the Shweli valley were Wa-tien and Chiang-chu, four miles southeast of 

On 18 June the 36th Division opened the attack on Wa-tien as the 116th 
and 130th Divisions arrived before Chiang-chu. While the 53d Army prepared 
to attack Chiang-chu, the 4th Infantry marched southward toward Ku-tung on 
the road to Teng-chung. Ku-tung fell on 19 June, and Wa-tien on the 20th. 
Both Chinese and Japanese converged on Chiang-chu. Luckily for the Chinese, 
monsoon rains that for a fortnight had greatly hampered air support now lifted. 
The transports quickly completed airdrops to each Chinese division, while 
fighter bombers pounded the Japanese around Chiang-chu. The Japanese 
yielded Chiang-chu on 22 June, and the 51st Fighter Group profited by the 
break in the weather to strafe and bomb a long column of Japanese with ani- 
mal transport caught on the trail to Teng-chung. 

With the fall of Chiang-chu, the Japanese had been forced to abandon the 
upper Shweli valley, and were now moving in some disorder toward Teng- 
chung over three excellent trails. In Chiang-chu, they left behind large quan- 
tities of ammunition and a few pieces of artillery, suggesting a disorganized 
withdrawal. One hundred and fifty dead Japanese were found in Chiang-chu 
itself; more than 300 Chinese gave their lives for the village. South of Chiang- 
chu, the Japanese hastily destroyed their ponton bridge to slow the Chinese 

On reaching the Chiang-chu- Wa-tien-Ku-tung line, XX Group Army had 
wrested 4,000 square miles from Japanese control in forty days of fighting. The 
advance had been made over the precipitous ranges of the Kaoli-kung Moun- 



tains in an almost constant rain, a downpour sometimes heavy, sometimes 
light, rarely abating, and always turning to fog and sleet in the higher altitudes. 
More than 150 coolie supply porters fell to their deaths from the narrow, slip- 
pery trails that snaked precariously over the mountains. 

Once solidly established in the Shweli valley, XX Group Army lost no time 
in ordering an advance on Teng-chung. The 2d Reserve Division plus elements 
of the 36th Division advanced southwest along the Ku-tung-Teng-chung trail. 
Advance patrols of the 198th Division proceeded directly south from Wa-tien 
while the rest of the division stayed to reorganize after its grueling advance 
through Ma-mien pass. The 53d Army was ordered to move from Chiang-chu 
and across the Shweli for an eventual concentration on the east side of Teng- 
chung. By 25 June General Wei received a personal order from the Generalis- 
simo to take Teng-chung. A few days later, XX Group Army, though delayed 
by the need to rebuild bridges over each of the swift mountain streams that 
crossed its advance, had pressed the Japanese rear guards back to the hills that 
surrounded Teng-chung at a distance of two to three miles from the formidable 
walled town itself. Meanwhile, the Fourteenth Air Force was trying to soften 
Teng-chung by daily attacks with bomb and machine gun. 

During the clearing of the Shweli valley and the approach to Teng-chung, 
Y-FOS personnel with XX Group Army sent back a number of critiques of 
that force in action, reports which suggested the Chinese advance might have 
gone more swiftly and cheaply had American techniques and advice not been 
disregarded by the Chinese on entering battle. Y-FOS observers wrote that 
Chinese regimental commanders could not ask directly for support from their 
attached artillery, but had to route their requests through division headquar- 
ters. When artillery support was granted, it was almost worthless. Targets were 
not bracketed, and delay between rounds was often as long as five minutes. Ar- 
tillery observers were sometimes two miles behind the front. The Chinese 
gunners disdained cover and concealment, drawing on themselves accurate Jap- 
anese counterbattery. Chinese pack artillery did not march in orderly fashion 
but straggled into position. Battery positions were occupied in daylight with 
individual pack sections arriving at half-hour intervals. The Chinese neglected 
to maintain their pieces, which quickly grew rusty during the rains. 

To the Americans, the Chinese seemed equally indifferent toward proper 
care and use of infantry supporting weapons. Chinese mortar crews dismissed 
their American-taught techniques. The firing batteries in rocket launchers were 
kept in place during the rains, which ruined them in twenty-four hours. The 
Chinese infantryman raised the hair on the Americans' heads by casually using 
the ring of the hand grenade to hang the weapon from his belt. Between all 
units, wire laying was bad. At night, an entire Chinese regiment would open 
up on a Japanese patrol. Ammunition was wasted endlessly, and weapons soon 
grew unserviceable from constant use and lack of maintenance. Such practices 
on a battlefield, far removed from supply depots and over 12,000 miles from 



the makers of critical parts, reduced the fighting power of General Wei's 
72,000 men far below what it should have been. Y-FOS personnel recommend- 
ed that a strict inspection system be established to make spot inspections and 
corrections. They urged that a high-ranking Chinese officer be present at every 
such inspection to follow through on recommendations and insure their per- 
formance. 30 

Gradually, the Chinese came to be more receptive to U.S. advice than they 
had been. In retrospect, it might seem that Wei's troops had appeared to accept 
U.S. teachings in the training centers only through Chinese courtesy, but with 
no notion of losing face by actually applying what they had heard from a for- 
eigner. Then came combat, and standing side by side American and Chinese 
officers could see what results the Chinese attitude yielded. The senior U.S. 
liaison officer with XX Group Army wrote: 

As a result of the lessons of this campaign, I believe that Commanders of all ranks in the 
20th Army Group [sic] are in the most receptive state of mind as pertains to American 
supervised training. The fact that American and Chinese officers stood side by side and 
watched excessive casualties pile up day after day, chiefly as a result of violations of proper 
tactical and technical procedure, furnishes a common ground of ideas for improved training. 
These same Chinese officers saw important objectives taken and held, at comparatively small 
cost, by troops following American training doctrines. There can be no doubt but that the 
American Liaison Team concept has been justified. The Chinese have acquired a new respect 
for the American Liaison Officer who shared with them the dangers and hardships of cam- 
paign, and a new confidence and understanding of the American training methods that 
proved their correctness in the final test— battle. 31 

The Southern Flank, 11 May-30 June 1944 

To insure its hold on the key Burma Road towns of Lung-ling and Mang 
shih, the 36th Division had occupied most of the larger villages lying south and 
west of Lung-ling. Lung-ling was vital to the Japanese because a fair road ran 
north from it to Teng-chung. If Lung-ling fell the whole Japanese position 
would be unhinged and would have to move many miles west. This would 
probably make Teng-chung untenable, while the momentum of the Chinese 
coming into Burma along the Burma Road might be enough to carry them 
into the vital area around Lashio. Lung-ling, therefore, was a prize worth striv- 
ing for. When the Chinese first struck, the 56th Division judged the greatest 
danger to be in the upper Shweli valley, and so weakened its southern flank to 
aid Colonel Kurashige. Therefore in the opening phases of their operation the 
two Chinese task forces of XI Group Army that were attempting to converge 
on Ping-ka from the north and south made excellent progress. 

Eleven miles below Ping-ka a task force of the 76th Division crossed the 

30 (1) Ltr with atchd rpts cited n. 28(4). (2) Rad CFB 18989, Ferris to Stilwell, 22 Jun 
44. Item 2641, Bk 7, JWS Personal File. ( 3) Y-FOS To urnal. 

31 Quotation from Col Buckley's ltr cited |n. 28(4). | 



CHINESE 2D ARMY ARTILLERYMEN prepare to fire their 75-mm.pack howitzer 
in the Ping-ka valley area. 

Salween on D Day, while one from the 88th Division crossed seven miles 
above. Three days after the crossings, the 76th Division met outposts of the 
1/146 and forced them back to the heights overlooking Ping-ka. Meanwhile 
the 88th Division from the north was fighting through a series of fortified vil- 
lages as it headed south to join the 76th Division. By 16 May thirteen villages 
were occupied in the area northeast of Ping-ka, but the Japanese, as they with- 
drew, received reinforcements. Strengthened by the 2/113 the Japanese did not 
attack the Chinese who were pursuing them, but moved south and hit the Chi- 
nese 228th Regiment south and west of Ping-ka. On 23 May the Chinese were 
forced to yield hill positions. Next day the Chinese 226th and 228th Regiments 
tried to force a way across the ridge which forms the southeastern edge of 
Ping-ka valley. Not waiting to co-ordinate their efforts with the 88th Division 
coming down from the north, and deprived of air support by inclement 
weather, the Chinese found their efforts futile. The 1/146 had carefully selected 
mortar and machine gun positions, and had cleared fields of fire by burning 
away the brush. Each Chinese frontal attack failed with heavy losses. 

After 25 May the 2d Army could not hope to co-ordinate its efforts with the 
88th Division since the latter was ordered to rejoin the 87th and New 28th 



Divisions (71st Army) in a drive on Lung-ling. Therefore, 2d Army assumed 
responsibility for the conduct of operations in the area Ping-ka-Hsiang-ta- 
Mang-shih. It ordered the 76th Division to bypass Ping-ka, leaving the 226th 
Regiment to besiege the valley. With only enough men to place forty to fifty 
soldiers on every mile of the twenty-four-mile semicircular front, the 226th 
Regiment remained before Ping-ka until the Japanese yielded the valley in late 
September. Tragedy and comedy alternated in the long three months' trial of 
endurance. The Chinese regiment faced a Japanese battalion, the 1/146, and 
the Japanese easily filtered through the thin Chinese line. When, at night, the 
Japanese found it hard to locate Chinese positions, they could stir the Chinese 
into revealing themselves by uttering weird cries and slapping their rifle butts. 
Then bursts of fire would spread back and forth across the front as the Chinese 
blazed away at the noises. But the Japanese paid for their tenacity. Cholera and 
malaria plagued the Ping-ka valley, and the Japanese twice had to reinforce the 
1/146. Finally, on 22 September, the Japanese sent a rescue column to evacuate 
Ping-ka and bring out what disease and malnutrition had left of a first-rate 
fighting team. 

On 1 June 2d Army brought two regiments of the 9th Division across the 
Salween and sent patrols of the 76th Division probing toward the Burma 
Road. The 9th Division isolated the Japanese in Ping-ka valley when it cut 
supply lines running from the road junction town of Hsiang-ta that lay be- 
tween Lung-ling and Ping-ka. Hopes of American liaison officers with the 2d 
Army that a speedy victory on the Salween front might be in the making 
soared when a unit of the 9th Division put a block across the Burma Road it- 
self on 9 June, at a point four miles south of Mang-shih. These hopes promptly 
foundered on the hidden rocks of Chinese Army politics. The 2d Army sus- 
pended its operations and complained bitterly that it was being discriminated 
against in supply. Investigating the charge, Y-FOS found that there was an 
old feud between Headquarters, XI Group Army, and Headquarters, 2d Army. 
American attempts to point out the results that might flow from holding a 
block on the Burma Road were not enough to close the breach. The 2d Army 
took its block off the Burma Road and contented itself with holding what it 
had and with brushing off Japanese patrols. 32 

list Army and the Fight for Lung-ling, 28 May-15 July 1944 

Aware that the 56th Division had withdrawn troops from its southern flank 
to meet XX Group Army, Gen. Wei Li-huang decided to exploit the situation 
by throwing in the rest of his forces. At the end of May he ordered the re- 
mainder of the 71st Army to cross the Salween below the Hui-tung Bridge to 

52 (1) Rpts cited ln. 23(2)1 (2) Y-FOS Journal. (3) Rpt, 2d Lt Raymond D. Ashman, Jul 
44, sub: Action of 226th Regt, 76th Div, at Ping-ka. AG (Y-FOS) 371.1, KCRC. (4) Japanese 
Study 93. 


seize Lung-ling while a containing force attacked the strong Japanese position 
on Sung Shan. Thus, hoping to gain an early victory over the 56th Division, 
Wei decided to commit his entire force, save the 8th Army then en route from 
the Indochina border, on three widely separated sectors. 

G-4, Y-FOS, had argued against the decision. questioned Sino-Amer- 
kan ability to sustain a twelve-division offensive, It pointed out that the aban- 
donment of the original concept of using small, highly mobile task forces and 
the decision to reduce every Japanese position in turn was placing a heavy bur- 
den on the long-neglccted Chinese services of supply. Moreover, the 27th 
Troop Carrier Squadron was complaining about having to support four Chi- 
nese armies on a hundred-mile front with but thirteen C-47's, suggested 
that attention should now be given to bringing forward replacements. The 
methods of attack to which the Chinese had reverted were steadily reducing the 
combat strength of their units. Replacing these losses was a major task requir- 
ing early planning and constant attention, for the nearest Chinese troop pool 
was H6 miles from Lung-ling and Chinese transport to the front was nor in 
shape to move forward a mass of replacements on short notice. The Fourteenth 
Air Force added its objections by sttessing the difficulty in giving air support 
to so many Chinese units with its few available aircraft. 33 

Stilwell, however, on learning of Wei's decision, joined Dorn in welcom- 
ing it. Current reports on the progress of the fighting at Myitkyina were most 
encouraging, and so, addressing Dorn by radio on 3 June, Stilwell directed him 
to encourage Wei. Stilwell was most anxious to join his forces with Wei's for 
he wished then to move thE Chinese Army in India and Wei's forces to east 
China to meet the Japanese threat. "Impress on all concerned the vital import- 
ance of getting forward on your front. I refer to what may happen soon in 
central and south China. It would be wise to establish early contact with 
X-Ray [Chinese Army in India], Then your boys could at least eat. You had 
better arouse them to realization of future possibilities. How are you doing?" H 

General Wei's decision to commit the 71st Army (less the 88th Division 
which had already crossed near Ping-ka) came just as a sudden and heavy rain 
caused the Sal ween to rise sharply. On 28 May the first elements commenced 
preparations for the ferrying operations west of Shih-tien, eight miles south of 
the Hui-tung Bridge. Three clays later, the Y-FOS liaison team with XI Group 
Army left Pao-shan to assist. Under the direction of Y-FOS engineers. 20,000 
troops reached the west bank of the Sal ween by 5 June. Unopposed by the Jap- 

51 ( 1) RpC Col "Walter S, Wood, Ct\ U.S. Ln Gp, XI Group Army, to Ha USAF CBi, Jan 45, 
sub; Salween Campaign, (HcreafteT Wood Report.) OCMH. (2) Memo, Col Reynolds Condon, 
G-4, Y-FOS, for Dorn, 2 Jul 44. sub: Supply of Opns. Y-FOS 1944 Hist Rpt. (3) Memo, Dorn 
for Stilwell, 23 May 44; Ltr. Col Clayton B. Claassen to Chennault, 17 Jul 44. sub; Rpt, Conf, 
Fourteenth AF Personnel with Dorn. SNF 35, (4) Memo. Dorn for Chennault, 13 Jul 44. AG 
(Y-FOS) 381, KCRC (5) Memo. CO 27th Tr Carrier Sq to CO T 6Ptb Composite Wine, 22 Jim 
44, sub: Air Supply Opns. AG ( Y-FOS} 373, KCRC. 

M 11) Rad CHC 1141, Stilwell to Dorn, 3 Jun 44. Item 198, Bit 6A, JWS Personal File. (2) 
Lc, Dorn to Ward, 2 Jun 52. OCMH. (3) Stilwell Diary, 2 Jun 44. 



anese, 71st Army climbed up and out of the Salween gorge and started north 
and west toward the Burma Road. Since its pack artillery was carried by coolies, 
but few rounds of ammunition were with the pieces. 

At La-meng village, four miles west of the Salween, the New 28th Division 
attacked the outposts of the Sung Shan Defense Unit, forcing it to abandon the 
village on 4 June and to fall back into the mountain fortress of Sung Shan. 
While the New 28th Division deployed to contain the 900 effective troops of 
the Sung Chan garrison, the 87th Division kept on toward Lung-ling. There 
it would meet the 88th Division which, supplied by air during its march from 
Ping-ka, had arrived outside the east gate of Lung-ling on 7 June. The Japanese 
still held onto the Burma Road east of Lung-ling, but as of 7 June the 87th 
Division had covered about two thirds of the distance along the Burma Road 
from the river to Lung-ling. The Chinese had to deal with small Japanese 
tanks, which had some moral effect but failed to halt the Chinese advance. In- 
deed, the latter had been fairly swift, for the Chinese had surprised the Jap- 
anese, and had been able to ease their supply problems by the capture of some 
Japanese rice depots. The scale of the fighting down the road to that point is 
suggested by the figure of known Japanese dead, 150 during the first week. 

The 88th Division deployed its troops along the eastern and southeastern 
heights overlooking Lung-ling and broke through the outer Japanese defenses 
on 8 June. That same day its sister division, the 87th, reached Lung-ling's 
North Gate. By sending the 26 1st Regiment off to the north, XI Group Army 
took control of the Man-lao Bridge on the Teng-chung-Lung-ling road, thus 
blocking that important Japanese supply route. " 

If the Chinese could take Lung-ling and break the center of the Japanese 
line on the Salween front, the five crack Chinese divisions under Stilwell and 
the twelve under Wei would be very near to meeting. The whole Japanese posi- 
tion in north Burma might crumble, and the blockade of China would then 
be broken. After a brief period for resting, refitting, and recuperating, the divi- 
sions of the Chinese Army in India could be flown to east China from the 
Myitkyina air strips to stiffen the defense of that area, if the Generalissimo, as 
Supreme Commander, China Theater, should so order, and if the local war 
area commanders should desire the presence of central-government troops. 36 
Wei's twelve divisions would be in position to cover this airlift. When the 
ground dried in October, a line of communications over existing and usable 
roads would be open from the Myitkyina-Mogaung area to Kunming. Over 
it, Wei's divisions could receive lend-lease equipment, and from Kunming, re- 
placements, always assuming the Chinese Ministry of War so desired. Then, 

35 (1) Wood Report. (2) Notes, Maj Herbert E. Feldman, U.S. Ln Off, New 28th Div, 1944. 
OCMH. (3) Japanese Study 93. 

36 General Merrill was convinced that Stilwell would have airlifted the Chinese 38th Division to 
east China. One may speculate that others would have followed, as they were to do in late 1944- 
early 1945. Interv with Merrill, 20 Apr 48. OCMH. 



if the military situation in Burma and the political situation in China permit- 
ted, units of the twelve divisions could be moved to east China, and there re- 
ceive a measure of logistical support made possible by taking north Burma. 
The intervention of any or all of Wei's divisions might have been ineffective, 
for their performance against one Japanese division does not, in retrospect, 
suggest they could have halted Hata's eleven divisions. But the five divisions 
under StilwelPs command had proved their efficiency, and the opportunity to 
move them to east China was a prize worth grasping. It was the only combat 
force Stilwell could offer to defend Chennault's airfields. Therefore, a victory 
at Lung-ling might offer the chance to save the day in east China. 

From the body of a dead Japanese officer, the Chinese took a map showing 
Lung-ling's defenses and the strength and composition of the garrison. Three 
5, 000- foot hills dominated the surrounding town and suburbs. A Japanese bat- 
talion held Lao Tung-po hill while a company of engineers and the garrison's 
few artillery pieces were on Shu Tung-po. The remaining 800 troops manned 
a central hill almost in the middle of town. 

Personally led by Gen. Sun Hsi-lien, XI Group Army commander, the 
youngest group army commander in the Chinese Army, 71st Army attacked 
the two outer heights from three sides on 9 June, in accord with the classic 
Chinese custom of leaving an escape route to a surrounded enemy.. 37 By 10 
June, the 87th Division had greatly reduced the volume of Japanese fire from 
Lao Tung-po, but despite some attacks by night the 88th was not able to 
silence a lone battery on Shu Tung-po. Inclement weather prevented tactical 
air support; 436 parachute loads dropped some 75-mm. ammunition for the 
Chinese, little enough to reduce the Japanese positions. Nor were 600 pack 
animals plying between the Shih-tien ferry over the Salween and Lung-ling 
able to keep pace with the expenditure of ammunition. Four days of unco- 
ordinated infantry attacks, with little artillery support, failed to carry the three 
mountains inside Lung-ling, and there was nothing to show for the heavy drain 
on the 71st Army's ammunition stocks. 

Despite numerous reports that the 56fh Division was gathering reinforce- 
ments for a counterattack, 71st Army did not consolidate and concentrate its 
available forces. Early on 14 June, 2d Army elements south of Lung-ling were 
attacked by Japanese patrols probing toward Lung-ling. That same morning, 
400 Japanese of the 113th Regiment who had been rushed south from Teng- 
chung crossed the Shweli west of the Chinese position at the Man-lao Bridge 
and attacked the outposts of the 261st Regiment. At dusk, 200 more Japanese 
ferried the Shweli, and at 1900 about one battalion of Japanese drove the Chi- 
nese off the bridge (which the latter did not destroy), and brought a small 
motorized column across the river. The Japanese inside Lung-ling counterat- 

37 (1) Wood Report. (2) Ltr, Col John K. Sells to authors, 1 May 47. OCMH. Colonel Sells 
was Commanding Officer, U.S. Liaison Team, 71st Army. (3) Dr. Ho Yung-chi in The Big Circle, 
page 130, explains the escape route device. 



tacked vigorously, and when 16 June ended, the 87th Division had been pushed 
back three miles from Lung-ling. 

The following day, Gen. Sun Hsi-lien ordered the 88th Division to aban- 
don its positions near Lung-ling and retire on a line with the 87th, eight miles 
to the northeast. Repeated attempts by American liaison personnel with the 
71st Army to learn how a small Japanese garrison could drive back a Chinese 
group army only brought embarrassed smiles from Chinese officers. Knowing 
by now what was prescribed by Chinese etiquette for facing the unpleasant, 
the Americans feared the worst. The Chinese finally related on 25 June that the 
26lst Regiment had bolted, and that the commanding general of the 87th Divi- 
sion had attempted suicide. Hopes somewhat revived that same day when re- 
ports came that the Honorable 1st Division (8th Army) 38 was arriving to rein- 
force, but in fact it was too late. So passed a brilliant opportunity; General 
Wei's attempt to exploit his initial successes by committing his reserves had 
been shattered by Sun's withdrawal before the counterattack of 1,500 Japanese. 
Meanwhile, in Burma, a major effort to take Myitkyina between 13 and 16 
June had brought no gains of great importance. 

When fuller details of the fighting around Lung-ling were available anger 
and annoyance spread from Y-FOS to the Generalissimo himself. Y-FOS per- 
sonnel considered the Chinese decision to withdraw from Lung-ling inexcus- 
able because XI Group Army had sent forward no reinforcements to meet the 
initial Japanese counterattacks. Of twenty-one battalions that XI Group Army 
had in the vicinity of Lung-ling on 14 June, only nine took part in the fight- 
ing. 39 The Japanese thus had been able to drive back 10,000 Chinese effectives 
by an attack with 1,500. In describing the defensive attitude of the 259th Regi- 
ment, as an example of the conduct that had cost the chance of a speedy break- 
through into Burma, one American liaison officer wrote: "From the time that 
we crossed the river until we reached Lung-ling, the regimental commander 
continually had his troops in the rear digging emplacements and trenches in 
the fear that they would have to retreat." 40 

When later reports on the Lung-ling battle filtered in to Dorn they were 
moderate in tone, and blamed faulty leadership and lack of training for the loss 
of the Chinese grip on the vital communications center. The Generalissimo 
was extremely angry when word reached him. Harsh orders from him, the ar- 
rival of the Honorable 1st Division, and a heavy raid by twenty-four B-25's on 
Lung-ling were all instrumental in halting the Japanese attempt to exploit 
their success. At Hwangtsoapa the Chinese forced the Japanese to fight for 

38 This division and one other, the Honorable 2d Division, received the honorific title of 
Honorable because they were made up of men who had been wounded and then returned to battle. 

,9 (1) Wood Report. (2) Y-FOS Journal. (3) Ltr cited n. 37(2). (4) Ltr, cited n. 34(2). 
(5) Japanese Officers' Comments, p. 32. (6) For details of the fighting at Myitkyina, see Chapter 
VI, above. 

40 Memo, Capt Frederick E. Van Tassell, CO, U.S. Ln Team, 259th Regt, for Dorn, 29 Jun 44. 
AG (Y-FOS) 319.1, KCRC. 



every foot of ground, though they were handicapped by especially inclement 
weather that interfered with air supply and air support. The Generalissimo in- 
tervened again with orders to General Wei to report all commanders who had 
shown themselves incompetent, and to spare no effort in eliminating Teng- 
chung and Sung Shan so that there might be a new effort from two directions 
against Lung-ling. 41 But the lost opportunity was past, and months of slow 
battering against Teng-chung and Sung Shan could not replace it. 


In mid-May, the Chinese crossed the Salween toward Burma. The cross- 
ing was unopposed, but hopes of a speedy break into Burma began to dwindle 
when the Chinese insisted on reducing Japanese strongpoints in turn, and dis- 
regarded the long-standing American advice that they should infiltrate through 
the scattered Japanese and move on into Burma. Nevertheless, considerable 
progress was made, and in late May Gen. Wei Li-huang decided to commit his 
central force to an attack straight down the Burma Road. After initial successes, 
the Chinese were thrown back from the key point of Lung-ling by a counter- 
attack of 1,500 Japanese and the middle of June found the Chinese with no 
hopes of a speedy break-through into Burma, while in east China the Jap- 
anese seemed to be moving at will. 

41 ( 1 ) Wood Re port. (2) Memo, Dorn for Ho Ying-chin, 5 Jul 44. AG (Y-FOS) 381. (3) 
Rad citec frl3t?71 


Facing the Command Problem 

The relationships of command within the Generalissimo's China Theater 
had not been thoroughly explored by the President and the War Department 
in concert since China Theater had been set up in January 1942, when the 
United States feared China might make a separate peace. What attention had 
been given to the command situation since then had been in the nature of 
specific responses to specific pressures from the Chinese or Chennault. The lack 
of harmony between the President and the War Department had not permitted 
continuing attention and close supervision. Therefore, no agency of the U.S. 
Government ever inquired as to why the Chinese had not been willing to set 
up an Allied staff for China Theater, as they had pledged themselves to do in 
1942, or, of course, sought to hold the Chinese to their promise. The issue of 
whether the Chinese would let Stilwell command any Chinese troops in China 
had been dropped by the Chinese as soon as he arrived in Chungking. The 
Soong-Stimson accord of January 1942, and the Generalissimo's reply to the 
inquiry of John J. McCloy, then Assistant Secretary of War, had implied such 
an intent on the part of the Chinese, but the U.S. Government had never 
pursued the matter. 1 

The impending Japanese offensive, threatening the Chinese Government 
with defeat, revived the command question. The Generalissimo's China 
Theater was an Allied theater, for two American air forces operated in it. Had 
all gone well in China Theater, probably the command situation would have 
stayed as it had for two years, with the question of Stilwell's exact powers and 
duties in that theater undefined. 

If the Generalissimo could hold east China, there would be no one to ques- 
tion his conduct of affairs. In 1937-38, when China's armies lost the Yangtze 
valley, the sea ports, and the key centers of north China, the loss could be 
ascribed to various causes beyond Chinese control, and since no American 
forces were involved, the U.S. Government could not concern itself with the 
quality of Chinese leadership. The events of 1944 followed on two years in 
which one group of American officers had predicted them, and threatened to 
affect the American effort in the Pacific. Moreover, they contrasted with the 

1 Stilwell's Mission to Cbinal Ch. V. 



unbroken chain of successes in north Burma, where Chinese troops under 
StilwelPs command had defeated some of the best units in the